Where things stand (2026-11-04)现状（2026-11-04）
Every row below carries its trust label exactly as cognition-os/03_TRUST_TABLE.md gives it, plus the revision-log changes R001–R131 and decision D30 folded in (08 revision 49; trust table read 2026-11-05 03:15 EST). Labels are not my judgement — they are the project's own. Numbers are quoted with their source.
下面每一行都按 cognition-os/03_TRUST_TABLE.md 给出的信任标签列出，并折叠了修订日志的变化 R001–R131 与决定 D30（08 修订版 49；信任表读取于 2026-11-05 03:15 EST）。标签不是我的判断——它们是项目自身的。数字均附有来源。
Established (trust = Trusted)已确立（信任 = Trusted）
Rows from 03_TRUST_TABLE.md whose label is Trusted. These are method/data-product rows, not physical detections of a hot M31 phase.
03_TRUST_TABLE.md 中标记为 Trusted 的行。这些是方法/数据产品行，而非热 M31 相的物理检出。
| Item项目 | Type类型 | Label (03)标签（03） | Number数字 | Source来源 | |
|---|---|---|---|---|---|
| 0.7 keV RGS detection0.7 keV RGS 检出 | Result结果 | TrustedTrusted | High高 | kT=0.74, O=1.93, Fe=0.89 (wilm)kT=0.74，O=1.93，Fe=0.89（wilm） | M31Center fiducial_medium_freeOFe_v2M31Center fiducial_medium_freeOFe_v2 |
| E\(\times\)P\(\times\)A region hierarchyE\(\times\)P\(\times\)A 区域层级 | Method方法 | TrustedTrusted | High高 | 3 orthogonal comparisons3 种正交对比 | claims.yaml; E006–E007claims.yaml；E006–E007 |
| WISE W1 as stellar mass proxyWISE W1 作为恒星质量代理 | Proxy代理 | TrustedTrusted | High高 | Rb=221″, Rd=850″, B/T=0.70Rb=221″，Rd=850″，B/T=0.70 | claims.yaml; Jarrett+2019 WXSCclaims.yaml；Jarrett+2019 WXSC |
| Variable-W1 q(a),PA(a)变 W1 q(a),PA(a) | Method方法 | TrustedTrusted | High高 | outer q=0.280, PA=37.7°, a=4–180′外部 q=0.280，PA=37.7°，a=4–180′ | claims.yaml; I007, E006claims.yaml；I007，E006 |
| Angr abundance table for re-fit重拟合用 angr 丰度表 | Method方法 | QuestionedQuestioned | Low低 | current fits use wilm — O/Fe need re-derivation当前拟合用 wilm——O/Fe 需重新推导 | 03; R2 pending (V5)03；R2 待定（V5） |
| 0.7 keV physical reality0.7 keV 物理真实性 | Hypothesis假设 | QuestionedQuestioned | Low低 | real M31 emission / MW foreground / O-norm degeneracy真实 M31 发射 / MW 前景 / O-norm 退化 | 00_AI_BRIEFING.md00_AI_BRIEFING.md |
| MWhothalo kT=0.7 (M31CGM)MWhothalo kT=0.7（M31CGM） | Cross-check交叉检验 | QuestionedQuestioned | Low低 | may be SrcApec-degenerate; not verified可能与 SrcApec 退化；未验证 | 00_AI_BRIEFING.md00_AI_BRIEFING.md |
| MOS 1.4–2.0 keV exclusionMOS 1.4–2.0 keV 排除 | Convention惯例 | SupersededSuperseded | High (historical)高（历史性） | now included (user decision 2026-07-19)现已包含（用户决定 2026-07-19） | 0303 |
| E034 old/young run2 completionE034 old/young run2 完成 | Provenance来源 | ResolvedResolved | High高 | old_only 11771.817, young_only 12541.776, mixed neverold_only 11771.817，young_only 12541.776，mixed 未完成 | R005; E034/E038R005；E034/E038 |
| Mixed old+young(W4dust)+constant混合 old+young(W4dust)+constant | Method方法 | RetiredRetired | High (linear forecast)高（线性预测） | corr(A_young,C)=\(-\)1.00; not identifiablecorr(A_young,C)=\(-\)1.00；不可辨识 | R006; E038R006；E038 |
| E034 “old beats young \(\Delta\)\(\chi^2\) 770”E034“old 胜 young \(\Delta\)\(\chi^2\) 770” | Result结果 | QuestionedQuestioned | Medium中等 | pedestal-laden W1 vs flat template带底座的 W1 vs 平坦模板 | R013; B026R013；B026 |
| obs/pred ratio 0.38→1.64obs/pred 比值 0.38→1.64 | Result结果 | ProvisionalProvisional | Medium中等 | 0.5–2 keV both sides; crosses 1 at 60–68′0.5–2 keV 双侧；在 60–68′ 处穿过 1 | R007; claims.yaml `theo-exp-obs-vs-predicted-profile`R007；claims.yaml `theo-exp-obs-vs-predicted-profile` |
| “not purely stellar” \(\chi^2\) reading“非纯恒星”\(\chi^2\) 读数 | Interpretation解释 | QuestionedQuestioned | Low低 | conditional errors; SrcApec–foreground degenerate条件性误差；SrcApec–前景退化 | R008 R008 |
| M31CGM v19 0.2 keV baselineM31CGM v19 0.2 keV 基准 | Calibration input校准输入 | ProvisionalProvisional | Medium中等 | 1.351e-15 \(\pm\) 9.6e-17; scatter 3.875e-16; +5.8% quality fields1.351e-15 \(\pm\) 9.6e-17；散射 3.875e-16；优质字段 +5.8% | R009; m31cgm_mwhalo_baseline.jsonR009；m31cgm_mwhalo_baseline.json |
| M/L_W1 = 0.6, L_X/M*M/L_W1 = 0.6，L_X/M* | Assumption假设 | Partly supportedPartly supported | Medium中等 | M/L=0.60 \(\pm\) 0.4 dex; L(0.5–2)/M*=(6–10)e27M/L=0.60 \(\pm\) 0.4 dex；L(0.5–2)/M*=(6–10)e27 | R010; Lit L01; Telford+2020R010；Lit L01；Telford+2020 |
| Minor-axis SE/NW sector profile小轴 SE/NW 扇区剖面 | Result结果 | Questioned (design)Questioned（设计） | Low低 | ObsID subsets; 23–72% area in other quadrantsObsID 子集；23–72% 面积在其他象限 | R011; E037R011；E037 |
| Major-axis NE/SW pilot a048_064大轴 NE/SW 试点 a048_064 | Result结果 | Provisional observationProvisional observation | Low低 | NE \(-\) SW \(-\)6.3\(\sigma\) (conditional); MWhalo +21% in SWNE \(-\) SW \(-\)6.3\(\sigma\)（条件性）；MWhalo 在 SW +21% | E037; R011E037；R011 |
| E033 branch B upper limitE033 分支 B 上限 | Result结果 | Withdrawn (not valid)Withdrawn（无效） | None无 | norm=8.206e-9, CI [0, 1.656e-8]; frozen sky — biased lownorm=8.206e-9，CI [0, 1.656e-8]；冻结天空——偏低 | R003; E033R003；E033 |
| E033 branch C “detection”E033 分支 C“检出” | Result结果 | Withdrawn (not valid)Withdrawn（无效） | None无 | kT=0.773, norm=3.390e-8, \(\Delta\)\(\chi^2\)=157; interval FAILEDkT=0.773，norm=3.390e-8，\(\Delta\)\(\chi^2\)=157；区间失败 | R004; E033R004；E033 |
| H1 MW stellar 0.7 keV foregroundH1 仅 MW 恒星 0.7 keV 前景 | Prediction预测 | Supported (conditional)Supported（条件性） | Medium中等 | (0.72–1.05)e-7 bright-removed; \(\leq\)1.60e-7 all-star(0.72–1.05)e-7 亮源移除；\(\leq\)1.60e-7 全恒星 | R014; E042R014；E042 |
| Disk nH frozen at 0.03e22圆盘 nH 冻结在 0.03e22 | Method assumption方法假设 | QuestionedQuestioned | Medium中等 | halo fields use 0.048–0.076e22; V1b (R116) then found that all 8 annuli prefer \(N_\mathrm{H}=0.0585\times10^{22}\) (\(\Delta\chi^2\) \(-\)27.8 to \(-\)218.6 at fixed dof), with the Milky Way halo norm \(+7.5\text{–}43\,\%\) and the SrcApec term \(-\)1.1 to \(+25.4\,\%\); the profile errors are still pending; a soft-calibration explanation is not excluded: a 2–3 % energy-dependent residual moves a fitted N_H by ~(0.9–2.5)e20 cm\(^{-2}\) (R124). The preference itself is now Questioned: with E < 0.5 keV removed the reduced refits prefer N_H 0.03 (\(\Delta\chi^2\) 3.79 and 16.73), so the 0.0585 value most likely reflects sub-0.5 keV calibration or soft-sky modelling rather than the column (R126)晕场用 0.048–0.076e22；随后 V1b（R116）发现全部 8 个环带都偏好 \(N_\mathrm{H}=0.0585\times10^{22}\)（固定自由度下 \(\Delta\chi^2\) \(-\)27.8 到 \(-\)218.6），银河晕 norm \(+7.5\text{–}43\,\%\)，SrcApec 项 \(-\)1.1 到 \(+25.4\,\%\)；剖面误差仍在等待；软波段校准的解释并未被排除：2–3 % 的能量依赖残余会把拟合的 N_H 移动 ~(0.9–2.5)e20 cm\(^{-2}\)（R124）。该偏好本身现在受质疑：移除 E < 0.5 keV 后，约化重拟合偏好 N_H 0.03（\(\Delta\chi^2\) 3.79 与 16.73），因此 0.0585 更可能反映 0.5 keV 以下的校准或软天空建模，而非柱密度本身（R126） | 03; V1/E035; R116, R12603；V1/E035；R116、R126 |
| MW-only foreground nH toward M31指向 M31 的仅 MW 前景 nH | Literature result文献结果 | Contradicted (magnitude)Contradicted（量级） | —— | \(\approx\)0.054–0.067e22, about twice 0.03\(\approx\)0.054–0.067e22，约为 0.03 的两倍 | R016 (Lit L02); 08 rev 14R016（Lit L02）；08 修订版 14 |
| South/SE baseline offsetSouth/SE 基准偏移 | Result结果 | Not robustNot robust | —— | \(-\)27.5% (producer mean) → \(-\)11.2% (p=0.237 with kT errors)\(-\)27.5%（生产者均值）→\(-\)11.2%（p=0.237，带 kT 误差） | R015; RT-05R015；RT-05 |
| obs/pred outer amplitudeobs/pred 外层幅度 | Result结果 | ProvisionalProvisional | Medium中等 | \(1.64\) (shipped baseline) to \(\approx 2.8\)\(1.64\)（已发布的基准）到 \(\approx 2.8\) | R017; RT-06R017；RT-06 |
| disk 0.2 keV MWhalo vs baseline圆盘 0.2 keV MWhalo vs 基准 | Result结果 | ProvisionalProvisional | Medium中等 | \(a=20'\) \(\approx 2.26\sigma\) borderline; 9 annuli consistent\(a=20'\) \(\approx 2.26\sigma\) 临界；9 个环带一致 | R018; RT-07R018；RT-07 |
| H1 emissivity caseH1 发射率情况 | Result结果 | —— | —— | bright-removed vs all-star undetermined亮源移除 vs 全恒星未确定 | R019; E043R019；E043 |
| stellar-floor discriminants恒星底板区分子 | Reference input参考输入 | SupportedSupported | Medium中等 | M32 H/S \(\approx 0.65\); L(0.5–2)/M* 5.4e27M32 H/S \(\approx 0.65\)；L(0.5–2)/M* 5.4e27 | R020; Lit L03R020；Lit L03 |
| hard/soft test E045硬/软测试 E045 | Result结果 | Provisional (hard-rich); Questioned → withdrawn (R048)临时性（硬质丰富）；受质疑 → 已撤回（R048） | Medium中等 | inner \(a \le 52'\) hard-rich H/S 1.9–3.0; soft floor 0.31e-15; QPB spectral-shape errors leave the five-annulus mean at 3.6–4.8\(\sigma\), a024_032 alone 1.85\(\sigma\) under a 10 % outside-band proxy (R052); RT-17 removes diffuse stray light (0.8–1.2 % of the excess), accepted-point SP covariance \(\leq\) 4.3 %, the \(\Gamma\) = 1.4 shape \(-\)4 to +5 %, and the off-axis QPB; the halo-CXB zero point and the omitted Au M / Cr / Mn / Fe K\(\alpha\) lines stay undetermined (limits as large as the effect); the halo zero is disfavoured by the outer disk (mean \(-\)0.94e-15: 5.3\(\sigma\) with the 20-field standard error, 1.3\(\sigma\) with one field's scatter); the SP shape needs the V23 refit (R058). V23 (R085) then freed the SP shape: it lowers \(\chi^2\) by 61.95 but moves the excess by \(+0.03 \pm 0.29\times10^{-15}\) (\(0.10\sigma\)), and all 112 added MOS-line norms stay at zero, so the SP shape and the omitted MOS lines no longer explain the excess; the spill/zero-point convention stays open (R078: 0.40–\(0.57 \times\) the Chandra value spill-corrected, 1.55–\(1.78 \times\) without the spill correction, a \(3.9\text{–}4.4\times\) convention difference). Like for like per K-band light: 1.59 \(\pm\) ~0.2\(\times\) the Chandra M31 quiet disk, 1.35 \(\pm\) 0.16 at the Chandra radii a016–a032, 1.84 \(\pm\) 0.15 across the ring a032–a056 (R056). The masks miss \(\leq\) ~¼ of the resolvable sources and they are not \(\Sigma\)★-shaped; faint LMXBs below the mask are 5–11 %; \(\approx\) 3e27 erg s\(^{-}\)\(^{1}\) M☉\(^{-}\)\(^{1}\) per solar mass stays unexplained (R054). The ring was measured before: Bogdán & Gilfanov 2008, 31.4 \(\pm\) 0.6e27 erg s\(^{-}\)\(^{1}\) L_K☉\(^{-}\)\(^{1}\) (R061). Faint X-ray binaries are a real but partial share: completeness weighting over 465 accepted fields (R096) gives an area-weighted 90 % removal completeness of 2.3–6.2e-15 erg cm\(^{-2}\) s\(^{-1}\) (2–4.5 keV) and an inner-six closure of 11–45 %; the energy-flux fix (R105) leaves the additive reading at 27–32 % and the LMXB-only reading at 4–5 %; the "HMXB" share is a model choice, not evidence of young origin (R102, R108)内 \(a \le 52'\) 硬质丰富 H/S 1.9–3.0；软底板 0.31e-15；粒子背景谱形误差使五环带均值停留在 3.6–4.8\(\sigma\)，a024_032 单独为 1.85\(\sigma\)（在 10 % 带外代理下）（R052）；RT-17 移除了弥散杂散光（超额的 0.8–1.2 %）、接受点 SP 协方差 \(\leq\) 4.3 %、\(\Gamma\) = 1.4 形状 \(-\)4 到 +5 % 以及离轴 QPB；晕-CXB 零点与遗漏的 Au M / Cr / Mn / Fe K\(\alpha\) 线仍未确定（限值与效应本身相当大）；晕零被外圆盘否定（均值 \(-\)0.94e-15：20 字段标准误下 5.3\(\sigma\)，单个字段散射下 1.3\(\sigma\)）；SP 形状需要 V23 的重拟合（R058）。随后 V23（R085）放开了 SP 形状：它使 \(\chi^2\) 降低 61.95，但把超额移动 \(+0.03 \pm 0.29\times10^{-15}\)（\(0.10\sigma\)），且全部 112 个新增 MOS 线 norm 保持在零，因此 SP 形状与遗漏的 MOS 线都不再解释该超额；溢出/零点约定仍然开放（R078：扣除溢出时为钱德拉值的 0.40–0.57 倍，不扣除时为 1.55–1.78 倍，两种约定在 H 中相差 3.9–4.4 倍）。按 K 光比光：1.59 \(\pm\) ~0.2\(\times\) 钱德拉 M31 安静圆盘，钱德拉半径 a016–a032 处 1.35 \(\pm\) 0.16，环带 a032–a056 处 1.84 \(\pm\) 0.15（R056）。掩护漏掉 \(\leq\) ~¼ 的可解析源，且它们不呈 \(\Sigma\)★ 形状；掩护下的暗 LMXB 占 5–11 %；每太阳质量 \(\approx\) 3e27 erg s\(^{-}\)\(^{1}\) M☉\(^{-}\)\(^{1}\) 仍未解释（R054）。环带此前已有测量：Bogdán & Gilfanov 2008，31.4 \(\pm\) 0.6e27 erg s\(^{-}\)\(^{1}\) L_K☉\(^{-}\)\(^{1}\)（R061）。暗弱 X 射线双星是真实但不完整的份额：对 465 个受理视场做完备度加权（R096），面积加权的 90 % 移除完备度为 2.3–6.2e-15 erg cm\(^{-2}\) s\(^{-1}\)（2–4.5 keV），内六个环带的闭合为 11–45 %；能量通量修正（R105）后，加性读法为 27–32 %、仅 LMXB 读法为 4–5 %；“HMXB”份额是模型选择，而不是年轻起源的证据（R102、R108） | R021, R047, R048, R052, R054, R056, R058, R061; E045R021、R047、R048、R052、R054、R056、R058、R061；E045 |
| WISE sky pedestalWISE 天空底座 | Bug impact错误影响 | NarrowedNarrowed | Medium中等 | 7.45 DN W1 mean-based; does not cause R0067.45 DN W1 均值底座；不导致 R006 | R022; B026R022；B026 |
| disk sessions off minimum圆盘会话非最小值 | QA质检 | QuestionedQuestioned | High (replayed)高（已重放） | 8/10 sessions lower \(\chi^2\) by 0.3–3.6 along SrcApec.kT8/10 个会话沿 SrcApec.kT 降低 \(\chi^2\) 0.3–3.6 | R023; B027R023；B027 |
| soft floor vs halo field软底板 vs 晕场 | Result结果 | Questioned / model-dependent受质疑 / 依赖于模型 | Medium中等 | \(s_0 = 3.11\times10^{-16}\) vs halo-field median \(2.93\times10^{-16}\) (within 6%); the link is model-dependent: s0/halo 0.65–0.71 with a soft \(\Sigma_\mathrm{SFR}\) term, 1.16–1.67 if the floor sits behind M31's own layer (R050, R051)\(s_0 = 3.11\times10^{-16}\) vs 晕场中位数 \(2.93\times10^{-16}\)（相差 6% 以内）；这一联系依赖于模型：带软 \(\Sigma_\mathrm{SFR}\) 项时 s0/晕场为 0.65–0.71，若底板位于 M31 自身气体层之后则为 1.16–1.67（R050、R051） | R024, R050, R051; RT-10R024、R050、R051；RT-10 |
| H1 MW stellar column independentH1 仅 MW 恒星柱密度独立复核 | Reference参考 | Supported (conditional)Supported（条件性） | Medium中等 | population mix \(-\)21%; completeness unrecoverable人口混合 \(-\)21%；完整性无法恢复 | R025; RT-09R025；RT-09 |
| halo-field 0.7 keV one level晕场 0.7 keV 一个水平 | Result结果 | Questioned (one common level; the detection itself stands)受质疑（一个共同水平；检出本身成立） | Medium中等 | The component is detected (R071, R079, R110, R112: q = 34.84, kT \(\le 0.80\) keV at 90 %, norm interval [3.15, 5.82]e-8 arcmin\(^{-2}\)), but its level is estimator-dependent: the v19 fit gives \(9.66 \pm 1.16e-8\) and the E036 fit 5.0 [3.56, 7.43]e-8, 2.3-2.7\(\sigma\) apart, and the registered levels span \(\times\)11.8 across fits of the same data (R093). The earlier single level (0.97 \(\pm\) 0.12e-7 at 8\(\sigma\), flat in M31 radius) is one of those readings, not the settled one该成分被检出（R071、R079、R110、R112：q = 34.84，90% 下 kT \(\le 0.80\) keV，norm 区间 [3.15, 5.82]e-8 arcmin\(^{-2}\)），但其水平依赖于估计器：v19 拟合给出 \(9.66 \pm 1.16e-8\)，E036 拟合给出 5.0 [3.56, 7.43]e-8，相差 2.3-2.7\(\sigma\)，且已注册水平在同一数据的拟合间跨度为 \(\times\)11.8（R093）。早先的单一水平（0.97 \(\pm\) 0.12e-7，8\(\sigma\)，在 M31 半径方向平坦）只是这些读法之一，并非定论 | R027, R071, R093, R110, R112; RT-11R027、R071、R093、R110、R112；RT-11 |
| inner hard excess vs PSF spill内硬超额 vs PSF 溢出 | Result结果 | Questioned (narrowed)Questioned（收窄） | Medium中等 | joint wing amplitude A = 0.70 [0.55, 0.85] of the King prediction; spill \(0.54\) (soft) / \(0.91\) (hard) e-15 at \(a=16\)–\(24'\); a shallower inner mask for unresolved background AGN is disfavoured — the predicted unresolved cosmic X-ray background is flat with no exposure trend, and no paper shows inward shallowing (R052)联合翼幅度 A = 0.70 [0.55, 0.85]（King 预测的比例）；在 \(a=16\)–\(24'\) 处溢出为 0.54（软）/ 0.91（硬）e-15；更浅内掩护下的未解析背景 AGN 不利——预测的未解析宇宙 X 射线背景是平的且无曝光趋势，也没有文献显示向内变浅（R052） | R035, R038, R052; V13d, V19R035、R038、R052；V13d、V19 |
| H1 residual stellar foregroundH1 残余恒星前景 | Result结果 | Supported (conditional)Supported（条件性） | Medium中等 | \(0.5\text{--}1.0\times\) bright-removed at XMM depth在 XMM 深度下为亮源移除的 0.5–1.0 倍 | R029; Lit L04R029；Lit L04 |
| MW 0.7 keV component toward M31指向 M31 的 MW 0.7 keV 成分 | Reference参考 | SupportedSupported | Medium中等 | HaloSat 3.28e-16 vs our 2.9e-16HaloSat 3.28e-16 vs 我们的 2.9e-16 | R030; Lit L06R030；Lit L06 |
| M31 unresolved hard emissivity / XLFM31 未解析硬发射率 / XLF | Reference参考 | —— | —— | M31 bulge X/K (4.5 / 3.6)e27; primary PDFs readM31 球状团 X/K (4.5 / 3.6)e27；已读原文 PDF | R031; Lit L05bR031；Lit L05b |
| Diffuse M31 disk 0.7 keV gas limit (stars + foreground subtracted)弥散 M31 圆盘 0.7 keV 气体上限（恒星 + 前景扣除后） | Result结果 | Provisional, conditional (R064 → R091 → R098; R131: the D1b SFR-shaped gas term not detected)临时性，条件性（R064 → R091 → R098；R131：D1b 的 SFR 形状气体项未检出） | Medium-low中低 | RC1 (R091): with a free fitted floor only the HI-shaped gas term is separable — coefficient \(3.47 \pm 1.56\times10^{-17}\) (\(2.2\sigma\)), so \(L_X(0.5\text{–}2\,\mathrm{keV}) < 3.8\times10^{37}\) erg/s inside \(a=56'\) and \(< 4.4\times10^{37}\) outside, about \(6\times\) below the R064 borrowed-floor limit. RT-27 (R098) reproduces the arithmetic exactly, but the term spans \(1.1\text{–}4.3\sigma\) across floor anchors, annuli, bands and cameras, so the limit is quoted as the envelope \(2.3\text{–}6.2\times10^{37}\) (inner) / \(2.7\text{–}7.1\times10^{37}\) erg/s (outer) with its floor anchor named, not as one number. Flat and SFR-shaped gas are degenerate with the floor and the young terms; the floor is anti-correlated with the HI term (\(-0.75\)); FG1, an independent floor field, decides. RT-38 (R131) then removed the SFR-shaped gas term of D1b altogether: adding the masked point-source count within 1′ as a covariate cuts its amplitude 1.3603 → 0.3673 (−73 %), the cluster-bootstrap SD 0.456 is \(\approx\) 21\(\times\) the naive SE, and the interval [0, 0.97]e−16 includes zero, so at 90 % that term is < 0.97e−16 \(\approx\) L < 1.1e37 erg/s over the footprint. The HI-shaped limit above is the one that standsRC1（R091）：在自由拟合底板下只有 HI 形状的气体项可分离——系数 \(3.47 \pm 1.56\times10^{-17}\)（\(2.2\sigma\)），因此 \(a=56'\) 以内 \(L_X(0.5\text{–}2\,\mathrm{keV}) < 3.8\times10^{37}\) erg/s、以外 \(< 4.4\times10^{37}\) erg/s，比 R064 借用底板的上限低约 \(6\) 倍。RT-27（R098）精确复现了算术，但该项随底板锚点、环带、波段与相机在 \(1.1\text{–}4.3\sigma\) 之间变动，因此上限以区间 \(2.3\text{–}6.2\times10^{37}\)（内）/ \(2.7\text{–}7.1\times10^{37}\) erg/s（外）引用并注明其底板锚点，而不是单一数字。平坦与 SFR 形状的气体与底板及年轻项退化；底板与 HI 项反相关（\(-0.75\)）；FG1（独立的底板视场）将决定。随后 RT-38（R131）完全去掉了 D1b 的 SFR 形状气体项：把 1′ 内的被掩护点源计数作为协变量加入后，其幅度从 1.3603 降到 0.3673（−73%），簇自助法的 SD 0.456 \(\approx\) 朴素 SE 的 21\(\times\)，区间 [0, 0.97]e−16 包含零，因此 90% 下该项 < 0.97e−16 \(\approx\) 在该足迹上 L < 1.1e37 erg/s。上面那条 HI 形状的上限才是站得住的 | R064, R091, R098; RC1; RT-27R064、R091、R098；RC1；RT-27 |
| one 0.7 keV foreground (I021)一个 0.7 keV 前景（I021） | Hypothesis假设 | one common level QUESTIONED (R093: the level is estimator-dependent — the v19 fit gives \(9.66 \pm 1.16\times10^{-8}\) and the E036 fit \(5.0 [3.56, 7.43]\times10^{-8}\), \(2.3\text{–}2.7\sigma\) apart, and registered halo levels span \(\times\)11.8 across fits of the same data; floor − halo is \(+1.64 \pm 1.41\times10^{-8}\) (\(1.2\sigma\)) with v19 and \(+6.30 \pm 1.52\times10^{-8}\) (\(4.2\sigma\)) with E036); the halo detection itself stands (R071, R079, R110, R112: q \(34.8\), kT \(\le 0.80\) keV at 90 %, norm \(3.2\text{–}5.8\times10^{-8}\) arcmin\(^{-2}\)); Milky Way origin Questioned, sharpened (R040; R057: with the near-disk published levels a pure constant is rejected, a compact component of core radius \(\approx\) 10 kpc is preferred, but 10 % shifts of the decomposition terms equal 52–92 % of the hot-level difference, so it is not identified as M31); the disk-floor link is model-dependent: s0/halo 0.65–0.71 with a soft \(\Sigma_\mathrm{SFR}\) term, 1.16–1.67 if the floor sits behind M31's layer (R050, R051)共同水平受质疑（R093：该水平依赖于估计器——v19 拟合给出 \(9.66 \pm 1.16\times10^{-8}\)，E036 拟合给出 \(5.0 [3.56, 7.43]\times10^{-8}\)，相差 \(2.3\text{–}2.7\sigma\)，而已注册的晕水平在同一数据的拟合间跨度为 \(\times\)11.8；floor − halo 在 v19 下为 \(+1.64 \pm 1.41\times10^{-8}\)（\(1.2\sigma\)），在 E036 下为 \(+6.30 \pm 1.52\times10^{-8}\)（\(4.2\sigma\)））；晕场检出一本身成立（R071、R079、R110、R112：q \(34.8\)，kT \(\le 0.80\) keV（90%），norm \(3.2\text{–}5.8\times10^{-8}\) arcmin\(^{-2}\)）；银河系起源受质疑且更锐化（R040；R057：带上已发表的近盘水平，纯常数被排除，偏好核半径 \(\approx\) 10 kpc 的紧致成分，但把分解项平移 10 % 相当于热水平差异的 52–92 %，因此它未被识别为 M31）；圆盘底板与晕场水平的联系依赖于模型：带软 \(\Sigma_\mathrm{SFR}\) 项时 s0/晕场为 0.65–0.71，若底板位于 M31 气体层之后则为 1.16–1.67（R050、R051） | Medium中等 | halo-field \(0.97 \pm 0.12\times10^{-7}\) \(\approx\) disk floor \(1.13 \pm 0.08\times10^{-7}\) \(\approx\) H1 0.72–1.05e-7; a cored M31 halo of 1.1–1.3e10 M☉ inside 150 kpc fits the 20 fields as well as a constant; R057: with the near-disk published levels a pure constant is rejected (\(\Delta\)\(\chi\)\(^{2}\) \(-\)22 to \(-\)28 across error sets) and a compact component of core radius \(\approx\) 10 kpc is preferred, but 10 % shifts of the cool/CXB/AGN terms equal 52–92 % of the hot-level difference, so the component is not identified as M31; R062 (interim, 5 of 18 fits): freezing the halo-field level into the outer fits drives the plasma term to its zero bound in a072_080 and a080_088 and the \(\chi\)\(^{2}\) rises past the pre-registered threshold in a064_072 and a080_088, so branch (C) "foreground over-subtracts" is indicated but not final; transplanting the halo level onto the outer disk gives an intercept C = \(-1.25 \pm 1.04\times10^{-16}\) (1.2\(\sigma\), the error the decision rule names) after the N_H/kT convention correction, so the arithmetic no longer decides (R053; the earlier \(4.2\sigma\) over-subtraction used conditional errors and was a dispatcher error)晕场 \(0.97 \pm 0.12\times10^{-7}\) \(\approx\) 圆盘底板 \(1.13 \pm 0.08\times10^{-7}\) \(\approx\) H1 0.72–1.05e-7；150 kpc 内 1.1–1.3e10 M☉ 的带核 M31 晕与常数拟合得一样好（20 个晕场）；R057：带上已发表的近盘水平，纯常数被排除（各误差组下 \(\Delta\)\(\chi\)\(^{2}\) \(-\)22 到 \(-\)28），偏好核半径 \(\approx\) 10 kpc 的紧致成分，但把 cool/CXB/AGN 项平移 10 % 相当于热水平差异的 52–92 %，因此该成分未被识别为 M31；R062（中期，18 个拟合中 5 个）：把晕场水平冻结进外层拟合后，等离子体项在 a072_080 与 a080_088 被压到零界，\(\chi\)\(^{2}\) 在 a064_072 与 a080_088 越过预注册阈值，因此分支 (C)“前景过度扣除”已被指示，但尚未最终确定；把晕场水平移植到外圆盘给出截距 C = \(-1.25 \pm 1.04\times10^{-16}\)（1.2\(\sigma\)，即决策规则指定的误差），这是 N_H/kT 约定修正之后的结果，因此算术不再决定问题（R053；早先 \(4.2\sigma\) 的过度扣除使用了条件性误差，是调度方错误） | R032, R040, R050, R051, R053, R057, R062; I021R032、R040、R050、R051、R053、R057、R062；I021 |
| V13 PSF spillV13 PSF 溢出 | Result结果 | QuestionedQuestioned | Medium中等 | spill not established; CCF tails 0.66\(\times\) King; wings absent in one ObsID溢出未确立；CCF 尾为 King 的 0.66 倍；一个 ObsID 中无翼 | R033; RT-13R033；RT-13 |
| faint-end stellar logN–logS暗端恒星 logN–logS | Reference参考 | NarrowedNarrowed | Medium中等 | L04 steep-slope bounds withdrawn; absolute residual not establishedL04 陡斜率下界已撤回；绝对残余未确立 | R034; Lit L07R034；Lit L07 |
| E040 decision ruleE040 决策规则 | Method方法 | Superseded (rule v2); the verdict is undetermined, power-limited (R122)Superseded（规则 v2）；裁定无法判定，受功效限制（R122） | High高 | rule v1 (A) passes by construction; v2 decides on the outer-three intercept C; after CORRECTION_03 the intercept from SrcApec alone is −\(1.32\sigma\) (HF1) and 0 (HF2), so the rule is undecided, and RT-35 showed the test is power-limited: the field scatter is 2–\(4 \times\) the whole halo range, so no per-annulus test at this precision can split halo from foreground (R115, R122)规则 v1 的 (A) 按构造即通过；v2 以外层三个环带的截距 C 判定；CORRECTION_03 之后仅 SrcApec 的截距为 −\(1.32\sigma\)（HF1）与 0（HF2），因此规则未决，而 RT-35 表明该检验受功效限制：视场散度是整个晕范围的 2–4 倍，故在此精度下任何逐环带检验都无法区分晕与前景（R115、R122） | R036; RT-12R036；RT-12 |
| solar-wind charge exchange太阳风电荷交换 | Reference参考 | NarrowedNarrowed | Medium中等 | 5 + 2 LU = 5.6e-16, 1.9\(\times\) the halo-field level; epoch-dependent; share untested5 + 2 LU = 5.6e-16，为晕场水平的 1.9 倍；随历元变化；占比未检验 | R037; Lit L09R037；Lit L09 |
| published near-disk hot level已发表的近盘热水平 | Reference参考 | NarrowedNarrowed | Medium中等 | 1.6–2.0\(\times\) our halo-field level (Kavanagh+2020); not yet like-for-like为我们晕场水平的 1.6–2.0 倍（Kavanagh+2020）；尚非同口径比较 | R039; Lit L10R039；Lit L10 |
| Paper I headline: M31 disk 0.7 keV needs no diffuse M31 gas论文 I 标题结论：M31 圆盘 0.7 keV 不需要弥散的 M31 气体 | Result (working conclusion, D27)结果（工作结论，D27） | Provisional, wording weakened (R080)临时性，措辞减弱（R080） | Medium中等 | Holds: current fits do not require diffuse M31 gas. Not established: a unique stars + flat Milky Way foreground decomposition (floor vs halo level \(1.1\sigma\) apart, model ratios 0.65–1.67, V7 HF1 over-subtracts the outer annuli). Wording: "consistent with stellar-like emission plus an extended foreground mixture". The independent test is FG1成立：当前拟合不需要弥散的 M31 气体。未确立：唯一的“恒星 + 平坦银河前景”分解（底板与晕场水平相差 \(1.1\sigma\)，模型比例 0.65–1.67，V7 HF1 在外环带过度扣除）。措辞：“与恒星状发射加上一个扩展的前景混合一致”。独立检验是 FG1 | I021, R050/R051, R062, R064, R071, R079, R080I021、R050/R051、R062、R064、R071、R079、R080 |
| PHAT \(\times\) XMM young/old per-mass contrastPHAT \(\times\) XMM young/old 每质量对比度 | Sensitivity灵敏度 | Provisional (R073, R081, R092 → R099; R121: the first age-resolved regression failed its physical calibration; R129: D1c formal only)临时性（R073、R081、R092 → R099；R121：首个按年龄分辨回归的物理校准失败；R129：D1c 仅形式结果） | Medium中等 | For the unresolved emission the expected contrast is 60–222\(\times\) (39–\(395 \times\) sweep); 2–7 keV 16.5–245\(\times\), below the ~\(1000 \times\) threshold; 0.5–2 keV 112–1680\(\times\), conditional on the old-soft emissivity and on diffuse SNRs. Only a ~\(1000 \times\) contrast is defensibly testable with the current counts (RT-21), but RT-34 removed F1's double coverage factor and the soft-band threshold is 16.43\(\times\) (\(33.84 \times\) with a floor nuisance), so the soft band is far more sensitive than the blanket figure suggested (R117). The first age-resolved regression returned a formal result whose soft QPB scale was 2.70 against a prior \(1 \pm 0.10\) (\(17\sigma\)) and whose deviance/dof was 3.59, so its per-age emissivity is not established (R121); D1b returned a formal, improved result with a certified response, but its physical gate still fails: the QPB/SP degeneracy is unresolved and the old-only fit gives \(\epsilon\) 3.63 (3.24–4.01) against M32 \(\approx\) 3.2, with free per-file floors returning \(\epsilon_{G4}\) = 0 (R125). D1c fixed the QPB per file from the detector corners but is formal only: its support rule removed 605/622 regions (13 regions and 2,926 counts left, 0/24 primary cases pass), and its full-sample row gives old \(\epsilon\) 1.99 [1.72, 2.25]e27 per formed M☉ with ratio 82, dev/dof 1.74 and QPB pull 8.4, so D1d was launched for pixel-level support and an absolute corner→FOV QPB (R129, R130). The soft QPB scale of 2.7 is now identified as the 1.4–2.0 keV Al K\(\alpha\) + Si K\(\alpha\) line band missing from the run6 continuum template, not soft protons (R127)对未解析发射，期望对比度为 60–222\(\times\)（归一化扫描 39–395\(\times\)）；2–7 keV 16.5–245\(\times\)，低于 ~1000\(\times\) 阈值；0.5–2 keV 112–1680\(\times\)，条件于旧软发射率与弥散 SNR。以当前计数仅 ~1000\(\times\) 对比度可辩护测试（RT-21），但 RT-34 移除了 F1 的双重覆盖因子，软波段阈值为 16.43\(\times\)（带底板 nuisance 时 33.84\(\times\)），所以软波段比笼统数字显示的灵敏得多（R117）。首个按年龄分辨回归给出了一个形式结果，但其软 QPB 尺度为 2.70，而先前为 \(1 \pm 0.10\)（\(17\sigma\)），deviance/dof 为 3.59，故其按年龄发射率未被确立（R121）；D1b 给出了有改进的形式结果并使用已认证的响应，但其物理门控仍失败：QPB/SP 退化未解决，仅 old 拟合给出 \(\epsilon\) 3.63（3.24–4.01），而 M32 \(\approx\) 3.2，自由的逐文件底板给出 \(\epsilon_{G4}\) = 0（R125）。D1c 用探测器角区逐文件固定 QPB，但只是形式结果：其支撑规则移除了 605/622 个区域（只剩 13 个区域、2,926 个计数，24 个主用例中 0 个通过），全样本行给出旧年龄 \(\epsilon\) 1.99 [1.72, 2.25]e27 每已形成 M☉、比值 82、dev/dof 1.74、QPB pull 8.4，因此启动了 D1d 做像素级支撑与绝对的角区→视场 QPB（R129、R130）。软 QPB 尺度 2.7 现已定位为 run6 连续谱模板中缺失的 1.4–2.0 keV Al K\(\alpha\) + Si K\(\alpha\) 线带，而非软质子（R127） | R073, R081, R092, R099, R117, R121, R123, R125, R127R073、R081、R092、R099、R117、R121、R123、R125、R127 |
| Hard 2–7 keV unresolved emissivity per disk mass圆盘质量对应的硬带 2–7 keV 未解析发射率 | Result结果 | Provisional, conditional (R083, R088: on the same-depth zero; not XRB-free; halo zero gives −4.6e27, XRB-corrected 5.8e27)临时性，有条件（R083、R088：在同深度零点上；非无 XRB；晕零点给出 −4.6e27，XRB 修正后 5.8e27） | Medium中等 | \(\epsilon\)_disk = 9.51 [7.85, 11.66]e27 erg/s/M☉; bulge unconstrained; the hard emission follows the disk. Faint XRBs from the New-ANGELS II XLFs give 4.6–16.5 % of the total hard (R084); completeness-weighted 11–45 % of the inner excess (R096), energy-corrected (R105): additive 27.3–32.4 %, LMXB-only 4.2–5.0 %; the old-template/LMXB overlap is checked by RT-32\(\epsilon\)_disk = 9.51 [7.85, 11.66]e27 erg/s/M☉；核球未约束；硬发射随圆盘。来自 New-ANGELS II XLFs 的暗弱 XRB 占总硬量的 4.6–16.5 %（R084）；完备度加权后占内超额的 11–45 %（R096），能量修正后（R105）：加性 27.3–32.4 %，仅 LMXB 4.2–5.0 %；old 模板/LMXB 重叠由 RT-32 检验 | R083, R084, R088, R096, R105R083、R084、R088、R096、R105 |
| S1 share package and the X1 region table: exposure mapsS1 共享包与 X1 区域表：曝光图 | Data product数据产品 | Exposure maps do not match the events (R109); X1b rebuilt them (R118); the legacy maps affect S1, X1 and F1 only (R119); the cause is now known — the S1 inventory preferred rpc/4.background (2.1filter GTI) while the events come from rpc/2.filter, and D30 fixes one GTI version for every product; D1d’s input audit found the same 4.background mask lineage inside S1b and queued CORRECTION_01 for the package (R130)曝光图与事件不匹配（R109）；X1b 已重建（R118）；传统图只影响 S1、X1 与 F1（R119）；原因现已查明——S1 清单优先选了 rpc/4.background（2.1filter 的 GTI），而事件来自 rpc/2.filter；D30 规定每个产品只用一种 GTI 版本；D1d 的输入审计发现 S1b 内部也有同样的 4.background 掩护谱系，并为共享包排入 CORRECTION_01（R130） | High高 | 0109270301 MOS1 events 26.5 ks vs map 11.5 ks; ONTIME/map 0.05–99 across 390 maps; 36 missing. Counts are exact. X1b rebuilt 426 maps on the event GTIs (MOS1 gate 0.9977) and the MOS QPB (622/826 boxes with a net rate; the hard band stays an upper limit). Every spectral-fit luminosity and the M31CGM spectrum-based surface brightness are unaffected, because they use the spectrum EXPOSURE and BACKSCAL, not the maps0109270301 MOS1 事件 26.5 ks 而图为 11.5 ks；390 张图的 ONTIME/map 为 0.05–99；36 张缺失。计数是精确的。X1b 在事件 GTI 上重建了 426 张图（MOS1 门控 0.9977）与 MOS QPB（622/826 个框有净速率；硬带仍为上限）。每个谱拟合光度与基于谱的 M31CGM 表面亮度不受影响，因为它们使用谱的 EXPOSURE 与 BACKSCAL，而非这些图 | R109, R118, R119, D30R109、R118、R119、D30 |
| fast_regions rc4 vs native v1.7.11 (D23 dual extraction)fast_regions rc4 vs 原生 v1.7.11（D23 双提取） | Validation验证 | annuli PASS in counts/EXPOSURE (36/36), QPB \(\pm 0.9\) % known rc4 normaliser defect; narrow sectors not testable: rc4 rejects DET POLYGON (12/12) (R100)环带在计数/EXPOSURE 上通过（36/36），QPB \(\pm 0.9\) % 是已知的 rc4 归一化缺陷；窄扇区不可测试：rc4 拒绝 DET POLYGON（12/12）（R100） | Medium中等 | Native v1.7.11 stays the production side; no speed-up is shown at \(\leq 10\) regions; the native sector cost is 1.61 h per task; POLYGON support has been requested from the tool owner原生 v1.7.11 仍是生产方；\(\leq 10\) 个区域时未显示加速；原生扇区成本为每任务 1.61 h；已向工具负责人请求 POLYGON 支持 | R055, R100R055、R100 |
| Inner-disk 2–7 keV excess: instrumental routes内圆盘 2–7 keV 超额：仪器途径 | Result结果 | SP shape and MOS lines disfavoured (R085)SP 形状与 MOS 线被否定（R085） | Medium中等 | SP shape shifts H by \(0.10\sigma\); 112 MOS-line norms stay at zero; the excess is \(1.08 \pm 0.42e-15\) in a016_024; faint XRBs supply 11–45 % of it after completeness weighting (R096; was 24–85 % at the hard cut, R084), mostly the faint "HMXB" term (the LMXB relabelling was tested by RT-29); the spill/zero-point convention is still open (R078)SP 形状使 H 移动 \(0.10\sigma\)；112 个 MOS 线 norm 保持为零；超额在 a016_024 为 \(1.08 \pm 0.42e-15\)；做完备度加权后暗弱 XRB 提供其中 11–45 %（R096；在硬切口处曾为 24–85 %，R084），主要是暗弱 "HMXB" 项（LMXB 重贴标签由 RT-29 检验）；溢出/零点约定仍然开放（R078） | R085, R096, R078R085、R096、R078 |
Source: cognition-os/03_TRUST_TABLE.md (all rows, read 2026-11-05 03:15 EST); cognition-os/07_REVISION_LOG.md (R001–R131); cognition-os/08_RUNNING.md revision 49.来源：cognition-os/03_TRUST_TABLE.md（全部行，读取于 2026-11-05 03:15 EST）；cognition-os/07_REVISION_LOG.md (R001–R131)；cognition-os/08_RUNNING.md 修订版 49。
Withdrawn & questioned claims (from 07)撤回与质疑的声明（来自 07）
| R-idR-id | Time时间 | Record记录 | Claim as recorded → what changed记录中的声明 → 变化 | Decision决定 | |
|---|---|---|---|---|---|
| R001 | 10:43 EST10:43 EST | E033E033 | E033 outcome Pending → complete (record only)E033 结果 Pending → complete（仅记录） | D19; back-filled from e033_report.mdD19；从 e033_report.md 回填 | |
| R002 | 10:43 EST10:43 EST | draft `old-young-template-run2-abandoned` vs E034草稿 vs E034 | consistent → conflicting (flagged)consistent → conflicting（已标记） | V4 settles; draft claims.yaml fixed after V4V4 解决；V4 后修正 draft claims.yaml | |
| R003 | 13:03 EST13:03 EST | E033 branch BE033 分支 B | Provisional → Withdrawn as valid upper limitProvisional → 作为有效上限撤回 | V2 full-freedom norm=0 null decidesV2 全自由度 norm=0 零假设决定 | |
| R004 | 13:03 EST13:03 EST | E033 branch CE033 分支 C | Proposed → Withdrawn as detectionProposed → 作为检出撤回 | V2 null fit decides; T-HALO07 reopenedV2 零假设拟合决定；T-HALO07 重开 | |
| R005 | 17:44 EST17:44 EST | E034 run2 completionE034 run2 完成 | Conflicting → resolved: E034 rightConflicting → 已解决：E034 正确 | claim reworded; 03 row closed声明重写；03 行已关闭 | |
| R006 | 17:44 EST17:44 EST | E034 / E038E034 / E038 | open → mixed with W4dust not identifiableopen → W4dust 混合不可辨识 | E034 mixed retired; V6 FUV mixed queuedE034 mixed 已退役；V6 FUV 混合排队 | |
| R007 | 17:51 EST17:51 EST | obs/vs-predicted profileobs/vs-predicted 剖面 | Active → Provisional (wording corrected)Active → Provisional（措辞已更正） | numerator is SrcApec SB in 0.5–2 keV, not 0.7–0.9 keV分子是 SrcApec SB 在 0.5–2 keV，而非 0.7–0.9 keV | |
| R008 | 17:51 EST17:51 EST | \(\chi^2\) “not purely stellar”\(\chi^2\)“非纯恒星” | Active → QuestionedActive → Questioned | profile-likelihood errors with foreground free become a requirement (V1b)带前景自由的 profile-likelihood 误差成为要求（V1b） | |
| R009 | 17:51 EST17:51 EST | MWhalo baseline M31CGMMWhalo 基准 M31CGM | Active → Provisional (selection-sensitive)Active → Provisional（对选择敏感） | baseline side asymmetry feeds V3; RT-05 tests it基准侧面不对称 feeding V3；RT-05 检验 | |
| R010 | 17:51 EST17:51 EST | assumptions M/L=0.6, SFR 0.3假设 M/L=0.6，SFR 0.3 | Assumed → Partly supportedAssumed → Partly supported | keep M/L 0.6 with 0.4 dex systematic保留 M/L 0.6，带 0.4 dex 系统误差 | |
| R011 | 18:45 EST18:45 EST | minor-axis SE/NW sector小轴 SE/NW 扇区 | Proposed → Questioned (design)Proposed → Questioned（设计） | spatially clipped extraction needed (PI item)需要空间剪切提取（PI 事项） | |
| R012 | 19:11 EST19:11 EST | E038 corrected forecastE038 修正预测 | Provisional → Provisional (lower bound)Provisional → Provisional（下界） | V6 must report marginal errorsV6 必须报告边缘误差 | |
| R013 | 19:11 EST19:11 EST | B026 WISE sky pedestalB026 WISE 天空底座 | Provisional → QuestionedProvisional → Questioned | V6 re-briefed: net templates, embedding checkV6 重新简报：净模板，嵌入检验 | |
| R014 | 19:16 EST19:16 EST | E042 H1 foreground predictionE042 H1 前景预测 | Proposed → Supported (conditional band)Proposed → Supported（条件性带） | V2 compares with band; run6 limit decidesV2 与带对比；run6 限值决定 | |
| R015 | 20:07 EST20:07 EST | R009 South/SE \(-\)27.5%R009 South/SE \(-\)27.5% | Provisional → Not robustProvisional → Not robust | no position-dependent baseline; recompute with kT–norm covariance无位置相关基准；用 kT–norm 协方差重算 | |
| R016 | 21:07 EST21:07 EST | Lit L02 MW-only N_HLit L02 仅 MW N_H | Questioned → Contradicted (magnitude)Questioned → Contradicted（量级） | MW-only \(\approx\)0.054–0.067e22 \(\approx\) twice 0.03; V1 bracket standsMW-only \(\approx\)0.054–0.067e22 \(\approx\) 0.03 的两倍；V1 括号保持 | |
| R054 R054 | 2026-11-02 16:47 EST2026-11-02 16:47 EST | V20 / Lit L17V20 / Lit L17 | What supplies the inner hard excess — missed point sources or faint LMXBs?什么供给内硬超额——被掩护漏掉的点源，还是更暗的 LMXB？ | Provisional → Provisional, narrowed: missed resolvable sources \(\leq\) ~¼ and not \(\Sigma\)★-shaped; faint LMXBs below the mask are 5–11 % of the excess; \(\approx\) 3e27 erg s\(^{-}\)\(^{1}\) M☉\(^{-}\)\(^{1}\) unexplainedProvisional → 临时性，收窄：被漏掉的可解析源 \(\leq\) ~¼ 且不呈 \(\Sigma\)★ 形状；掩护下的暗 LMXB 占超额的 5–11 %；\(\approx\) 3e27 erg s\(^{-}\)\(^{1}\) M☉\(^{-}\)\(^{1}\) 未解释 | |
| R055 R055 | 2026-11-02 16:47 EST2026-11-02 16:47 EST | V12 stage AV12 阶段 A | Can fast_regions replace native mos-spectra?fast_regions 能否替代原生 mos-spectra？ | R041 → stage A result: object spectra identical, QPB differs up to 0.9 %; replacement neither refuted nor qualifiedR041 → 阶段 A 结果：对象谱一致，QPB 差异达 0.9 %；替代方案既未被证伪也未被认可 | |
| R056 R056 | 2026-11-02 17:46 EST2026-11-02 17:46 EST | V21 / Lit L18V21 / Lit L18 | Is the inner hard emission above Chandra's M31 value per K-band light, like for like?按 K 光比光，内硬辐射是否高于钱德拉的 M31 值，同口径比较？ | R046/R054 → Provisional, narrowed and localised: 1.59 \(\pm\) ~0.2\(\times\) B10's quiet disk overall; 1.35 \(\pm\) 0.16 at B10's radii (a016–a032); 1.84 \(\pm\) 0.15 across the ring (a032–a056)R046/R054 → 临时性，收窄并局部化：比 B10 安静圆盘高 1.59 \(\pm\) ~0.2 倍整体；B10 半径处 (a016–a032) 1.35 \(\pm\) 0.16；环带处 (a032–a056) 1.84 \(\pm\) 0.15 | |
| R057 R057 | 2026-11-02 17:46 EST2026-11-02 17:46 EST | RT-16RT-16 | After RT-14's mass correction, is the one 0.7 keV level a Milky Way foreground or M31's halo?RT-14 的质量修正后，单一 0.7 keV 水平是银河系前景还是 M31 的晕？ | R040 → still Questioned, sharpened: pure constant rejected (\(\Delta\)\(\chi\)\(^{2}\) \(-\)22 to \(-\)28); compact r_c \(\approx\) 10 kpc preferred but decomposition systematics equal 52–92 % of the hot-level difference; not identified as M31R040 → 仍受质疑且更锐化：纯常数被排除（\(\Delta\)\(\chi\)\(^{2}\) \(-\)22 到 \(-\)28）；偏好核半径 r_c \(\approx\) 10 kpc 的紧致成分，但分解系统误差等于热水平差异的 52–92 %；未被识别为 M31 | |
| R058 R058 | 2026-11-02 23:36 EST2026-11-02 23:36 EST | RT-17RT-17 | Is the inner hard excess an instrumental or zero-point artefact?内硬超额是仪器效应还是零点伪迹？ | Provisional → Provisional, narrowed: four instrumental channels removed; zero point and omitted MOS lines undetermined; SP shape needs a refit; halo zero disfavoured by outer disk临时性 → 临时性，收窄：移除了四个仪器通道；零点和遗漏的 MOS 线未确定；SP 形状需要重拟合；晕零被外圆盘否定 | |
| R059 R059 | 2026-11-02 23:36 EST2026-11-02 23:36 EST | V6 / E039 pilotV6 / E039 试点 | Does a young (FUV) template or a constant enter the 0.7 keV emission in the a032_040 pilot?年轻（FUV）模板或常数项是否进入 a032_040 试点中的 0.7 keV 辐射？ | Q2 Questioned → new pilot result: old + FUV beats old + constant by \(\Delta\)\(\chi\)\(^{2}\) \(-\)7.34; A_young/A_old = 0.32 (degenerate); constant at its bound 0, < \(1.33 \times\) s0 after the \(\times 501\) unit correctionQ2 受质疑 → 新试点结果：old + FUV 胜过 old + constant，\(\Delta\)\(\chi\)\(^{2}\) \(-\)7.34；A_young/A_old = 0.32（退化）；常数项在其界限 0，\(\times\)501 单位修正后 < \(1.33 \times\) s0 | |
| R060 R060 | 2026-11-02 23:36 EST2026-11-02 23:36 EST | V14 / Lit L09V14 / Lit L09 | Does solar-wind charge exchange drive the halo-field 0.7 keV level?太阳风电荷交换是否驱动晕场 0.7 keV 水平？ | untested → first screening: no SWCX signature in halo level or scatter; disk/halo solar-phase mismatch is real but small; magnetospheric scenario moves a transplant \(-\)3 to \(-\)11 %未检验 → 首次筛查：晕场水平或散射中无 SWCX 签名；圆盘/晕场太阳相位差异真实但很小；磁层场景使移植变化 \(-\)3 到 \(-\)11 % | |
| R061 R061 | 2026-11-02 23:36 EST2026-11-02 23:36 EST | Lit L19Lit L19 | Has unresolved hard emission on M31's 10-kpc ring been measured before?M31 10 kpc 环带上的未解析硬辐射此前是否已被测量？ | R056 → context, labels unchanged: Bogdán & Gilfanov 2008 found 31.4 \(\pm\) 0.6e27 erg s\(^{-}\)\(^{1}\) L_K☉\(^{-}\)\(^{1}\) on two ring ellipses, 7–9\(\times\) the bulgeR056 → 上下文，标签不变：Bogdán & Gilfanov 2008 在两个环带椭圆上测得 31.4 \(\pm\) 0.6e27 erg s\(^{-}\)\(^{1}\) L_K☉\(^{-}\)\(^{1}\)，是核球的 7–9 倍 | |
| R062 R062 | 2026-11-03 00:45 EST2026-11-03 00:45 EST | V7 interim / expert review T2V7 中期 / 专家评审 T2 | Do the first E040 refits transplant the halo-field 0.7 keV level onto the outer disk?首批 E040 重拟合是否把晕场 0.7 keV 水平移植到外圆盘？ | R053 → interim, not final: branch (C) 'foreground over-subtracts' indicated; I021's label unchanged until V7 finishesR053 → 中期，非最终：分支 (C)“前景过度扣除”已被指示；I021 的标签在 V7 完成前不变 | |
| R063 R063 | 2026-11-03 00:45 EST2026-11-03 00:45 EST | expert review (Fable)专家评审（Fable） | Is M31HotISM done, what next?M31HotISM 是否完成，下一步是什么？ | goals not reached; 7 of 9 tasks accepted, 2 to the PI; T1 upper-limit packet accepted; T7/T8 to the PI; weak spot 2 rejected目标未达成；9 项任务中 7 项被接受，2 项提交 PI；T1 上限资料包被接受；T7/T8 提交 PI；弱点 2 被拒 | |
| R064 R064 | 2026-11-03 02:06 EST2026-11-03 02:06 EST | T1 upper-limit packet (opus)T1 上限资料包（opus） | Diffuse M31 0.7 keV gas limit弥散 M31 0.7 keV 气体上限 | 'no diffuse M31 gas required' → no number: 3\(\sigma\) L_X(0.5–2 keV) < 2.4e38 (a \(\leq 56\)′), < 1.6e38 (a \(\geq 56\)′) with a 0.4 dex floor error; Mineo-level gas disfavoured 2.2–\(3.2\sigma\) / excluded 3.8–\(6.4\sigma\)“不需要弥散 M31 气体” → 无数字：3\(\sigma\) L_X(0.5–2 keV) < 2.4e38 (a \(\leq 56\)′)，< 1.6e38 (a \(\geq 56\)′)，带 0.4 dex 底板误差；Mineo 级气体被否定 2.2–\(3.2\sigma\) / 排除 3.8–\(6.4\sigma\) | |
| R065 R065 | 2026-11-03 02:06 EST2026-11-03 02:06 EST | Lit L21Lit L21 | Diffuse hot-gas limits, per-SFR expectation弥散热气上限，按 SFR 的期望值 | Mineo-level gas would be seen → benchmark uncertain \(\times\)10Mineo 级气体本应被看见 → 基准不确定约 10 倍 | |
| R066 R066 | 2026-11-03 02:06 EST2026-11-03 02:06 EST | T5 sector-campaign sizingT5 扇区运动规模估计 | Sector campaign sizing before V8bV8b 之前的扇区运动规模估计 | 452 tasks → informative subset; minor axis cannot detect < ~86 % (untestable)452 项任务 → 有信息量的子集；小轴无法检测 < ~86 %（不可测试） | |
| R067 R067 | 2026-11-03 02:06 EST2026-11-03 02:06 EST | T6 V6b identifiability ruleT6 V6b 可辨识性规则 | V6b subset, rule and injection designV6b 子集、规则与注入设计 | CHALLENGE accepted: two pilot rows are not template rows; recommended subset F (a032_040 + a040_048 + a080_096)CHALLENGE 被接受：两个试点行不是模板行；推荐子集 F (a032_040 + a040_048 + a080_096) | |
| R068 R068 | 2026-11-03 02:06 EST2026-11-03 02:06 EST | Lit L22 / PHAT productsLit L22 / PHAT 产品 | Age-resolved X-ray emissivity, PHAT SFHs按年龄分辨的 X 射线发射率，PHAT 恒星形成历史 | open → priors in hand (narrowed): old-population L_X/M = (6–10)e27; falls ~3 dex from 10 Myr to 10 Gyropen → 先验已在手中（收窄）：旧种群 L_X/M = (6–10)e27；从 10 Myr 到 10 Gyr 下降约 3 dex | |
| R069 R069 | 2026-11-03 03:54 EST2026-11-03 03:54 EST | Lit L20Lit L20 | MOS instrumental lines 2–7 keV and ESAS QPBMOS 仪器线 2–7 keV 与 ESAS QPB | RT-17 → open → narrowed, not closed: Au M\(\alpha\) 2.123, Cr K\(\alpha\) 5.41, Fe K\(\alpha\) 6.40 keV are inside mos-back QPB if the code follows the cookbook (unverified); V23 tests directlyRT-17 → open → 收窄，未关闭：Au M\(\alpha\) 2.123、Cr K\(\alpha\) 5.41、Fe K\(\alpha\) 6.40 keV 若代码遵循烹饪手册则在 mos-back QPB 内（未验证）；V23 将直接检验 | |
| R070 R070 | 2026-11-03 03:57 EST2026-11-03 03:57 EST | RT-18 run6 product acceptanceRT-18 run6 产品验收 | Annulus fits rest on run6 products?环带拟合是否建立在 run6 产品之上？ | premise withdrawn: none of the 302 accepted spectra comes from run6 — they bind three v1.7.0 campaigns; run6 has 110 failed tasks and only 4 of 8 products per task前提已撤回：302 个受理谱中没有一条来自 run6——它们绑定三个 v1.7.0 运动；run6 本身有 110 个失败任务且每个任务只保留 8 个产品中的 4 个（图像被省略） | |
| R071 R071 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | V2 / E036V2 / E036 | halo-field SrcApec kT profile + full-freedom null晕场 SrcApec kT 剖面 + 全自由度零假设 | none -> DETECTED (Provisional; norm interval pending in V2b)无 -> 检出（临时性；norm 区间待 V2b） | 13-point grid under C's free set (dof 748, two starts, start a kept everywhere): best \(\chi^2\) 1253.536 at kT 0.76; null SrcApec.norm = 0 (dof 749): 1275.962 (start a; start b 1322.565). \(q_{\max} = 22.43 > 12.46\) -> detected (local \(p \approx 10^{-6}\), \(\times 13\)). kT 0.70 within \(\Delta\chi^2\) 1.35 (< 2.706); 90% kT 0.68–0.82. Norm 3.64e-8 at 0.70 (E033 C had 3.39e-8 at 0.773). Start b ~50 \(\chi^2\) worse in both null and grid: local minima, same start wins bothC 的自由集下 13 点网格（dof 748，两个起始值，始终保留起始 a）：最佳 \(\chi^2\) 1253.536 处 kT 0.76；零假设 SrcApec.norm = 0（dof 749）：1275.962（起始 a；起始 b 1322.565）。\(q_{\max} = 22.43 > 12.46\) -> 检出（局部 \(p \approx 10^{-6}\)，\(\times 13\)）。kT 0.70 在 \(\Delta\chi^2\) 1.35 以内（< 2.706）；90% kT 0.68–0.82。0.70 处 norm 3.64e-8（E033 分支 C 在 0.773 处为 3.39e-8）。起始 b 在零假设与网格中都差 ~50 的 \(\chi^2\)：局部最小值，同一起始值赢得两者 |
| R072 R072 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | RT-19RT-19 | V6 pilot young (FUV) termV6 试点年轻（FUV）项 | Provisional -> Weakened临时性 -> 减弱 | \(\Delta\chi^2\) +7.34 reproduces from the saved sessions, but A_young/A_old = \(0.325 \pm 0.346\) (\(0.94\sigma\), one-sided 95% lower −0.24); the gain rests on 2.08 effective spectra (top-3 = 84.5%, 21 of 43 worsen); old/young templates correlated (r 0.56, ObsID-cluster 95% CI 0.23–0.81). C \(\geq 0\) bound, frozen N_H and MW 0.7 keV halo cannot explain it; the FUV pedestal and sky subtraction stay unverified; young kT is tied to old, so no spectral leverage. Even 316 spectra would give ~\(2.5\sigma\)\(\Delta\chi^2\) +7.34 从保存的会话中复现，但 A_young/A_old = \(0.325 \pm 0.346\)（\(0.94\sigma\)，单侧 95% 下限 −0.24）；增益仅来自 2.08 个有效谱（前 3 个占 84.5%，43 个中有 21 个变差）；old/young 模板相关（r 0.56，ObsID 聚类 95% CI 0.23–0.81）。C \(\geq 0\) 界限、冻结 N_H 与 MW 0.7 keV 晕都无法解释它；FUV 底座与天空减法仍未验证；年轻 kT 与 old 绑定，故无谱杠杆。即使 316 个谱也只有 ~\(2.5\sigma\) |
| R073 R073 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | F1F1 | PHAT SFH design matrix and count forecastPHAT 星形成历史设计矩阵与计数预测 | open -> narrowed开放 -> 收窄 | Age grouping G4 is well conditioned (condition number 2.9–3.6, max VIF 2.3–3.7). With present MOS + pn counts only a \(\geq\)\(100 \times\) young/old contrast is detectable at \(3\sigma\) in 0.8–2 keV (median model: ratio 100 − \(3\sigma\) = 31.5; \(10 \times\) and \(30 \times\) not). Old \(\psi\) conditionally to 9.2% / 4.2% (two soft bands). A single old emissivity \(\times\) PHAT mass fails the radial profile: \(\chi^2\)/dof 24.8–46.8 (BASTI 0.8–2 keV 228/7). Particle and SP background not propagated; pn ARF representative年龄分组 G4 条件良好（条件数 2.9–3.6，最大 VIF 2.3–3.7）。以当前 MOS + pn 计数，仅 \(\geq 100\)\(\times\) 的 young/old 对比度在 0.8–2 keV 处可被 \(3\sigma\) 检出（中位模型：比值 100 − \(3\sigma\) = 31.5；10\(\times\) 和 30\(\times\) 不可）。old \(\psi\) 有条件地约束到 9.2% / 4.2%（两个软波段）。单 old 发射率 \(\times\) PHAT 质量无法通过半径剖面：\(\chi^2\)/dof 24.8–46.8（BASTI 0.8–2 keV 228/7）。粒子与 SP 背景未传播；pn ARF 具有代表性 |
| R074 R074 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | RT-18bRT-18b | v1.7.0 accepted inputsv1.7.0 已受理输入 | assumed -> verified, one bounded systematic假设 -> 已验证，有一个界定的系统误差 | 302/302 quartets and receipts hash-matched, EXPOSURE/BACKSCAL to 1e-6, no repeated or time-overlapping ObsIDs, annuli disjoint. Material: the two MOS cameras of an ObsID cover different sky parts of an annulus (area ratio median 1.3–2.3\(\times\), max 9.7\(\times\); hot-band SB differs median 11–21%), which shifts the shared per-arcmin\(^{-2}\) SrcApec.norm by ~3–9% (to ~35% at a016_024 p90); CXB_peg less302/302 个四元组与收据哈希匹配，EXPOSURE/BACKSCAL 匹配至 1e-6，无重复或时间重叠的 ObsID，环带不重叠。显著：同一 ObsID 的两台 MOS 相机覆盖环带的不同天空部分（面积比中位数 1.3–2.3\(\times\)，最大 9.7\(\times\)；热带 SB 中位数差 11–21%），这把共享的每平方角分（arcmin\(^{-2}\)）SrcApec.norm 移动约 3–9%（a016_024 p90 处约 35%）；CXB_peg 较小 |
| R075 R075 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | Lit L23Lit L23 | M31 bulge stellar populations and massM31 核球恒星族群与质量 | open -> priors in hand开放 -> 先验已在手 | Bulge > 70% older than 5 Gyr, mean 12–13 Gyr, [Fe/H] \(\approx\) +0.3 (Saglia+2018, Dong+2018); ~5% of bulge mass at ~1 Gyr; < 500 Myr only in the central 130″, none < 10 Myr. Inner disk mean ~8.5 Gyr. Bulge M★ \(\approx 3.1e10\) M☉ (Blaña Díaz+2018, Υ3.6 0.72), disk ~5.9e10; M/L_V bulge 4.4–4.7, disk 3.0. DERIVED: \(3.1e10 \times 8e27\) \(\approx 2.5e38\) erg/s old-star soft L_X for the bulge, the size of its observed diffuse emission核球 > 70% 年龄超过 5 Gyr，平均 12–13 Gyr，[Fe/H] \(\approx\) +0.3（Saglia+2018，Dong+2018）；约 5% 的核球质量在 ~1 Gyr；< 500 Myr 仅在中心 130″ 内，无 < 10 Myr。内圆盘平均 ~8.5 Gyr。核球 M★ \(\approx 3.1e10\) M☉（Blaña Díaz+2018，Υ3.6 0.72），圆盘 ~5.9e10；M/L_V 核球 4.4–4.7，圆盘 3.0。推导：\(3.1e10 \times 8e27\) \(\approx 2.5e38\) erg/s 核球旧恒星软 L_X，与其观测弥散发射规模相当 |
| R076 R076 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | Lit L24Lit L24 | EPIC-pn for faint diffuse emission用于微弱弥散发射的 EPIC-pn | open -> systematics listed开放 -> 系统误差已列出 | ESAS pn-back QPB per quadrant from corner + FWC data with OOT removed; Au/Cr/Fe weak lines included, Al K\(\alpha\) and Cu K\(\alpha\)/K\(\beta\) (8.03/8.90 keV, pn-only, hole under the electronics) excised; OOT fraction Full Frame 6.3%, Extended Full Frame 2.3%; SP residual per camera; MOS above pn by ~5–7% in 2–7 keV, 0–3% below 1 keV (IACHEC)ESAS pn-back QPB 由角 + FWC 数据按象限推导，OOT 已移除；Au/Cr/Fe 弱线已包含，Al K\(\alpha\) 与 Cu K\(\alpha\)/K\(\beta\)（8.03/8.90 keV，pn 独有，电子下方有空洞）已剔除；OOT 比例 Full Frame 6.3%，Extended Full Frame 2.3%；每相机 SP 残留；MOS 在 2–7 keV 中比 pn 高约 5–7%，在 1 keV 以下高 0–3%（IACHEC） |
| R077 R077 | 2026-11-03 09:54 EST2026-11-03 09:54 EST | Lit L25Lit L25 | log-normal temperature distributions对数正态温度分布 | open -> no (narrowed)开放 -> 否（收窄） | MW CGM log-normal widths \(\sigma\)(log T) \(\approx 0.20\)–0.23 dex; galaxies (M33, NGC 253, MW halo) are fitted with 0.2 + 0.6–0.7 keV pairs and CCD spectra cannot tell a continuous distribution from 2T. DERIVED: a log-normal centred at 0.2 keV with \(\sigma\) 0.23 dex has < 1% of its emission measure above 0.7 keV, so the pair needs a separate hot phase or \(\sigma\) ≳ 0.3 dex. M31's own 0.1–0.3 keV ISM is supported (Kavanagh+2020, Zhang+2019)MW CGM 对数正态宽度 \(\sigma\)(log T) \(\approx 0.20\)–0.23 dex；星系（M33、NGC 253、MW 晕）被拟合为 0.2 + 0.6–0.7 keV 双温，CCD 谱无法区分连续分布与 2T。推导：中心于 0.2 keV、\(\sigma\) 0.23 dex 的对数正态分布，其发射度中高于 0.7 keV 的部分 < 1%，因此该双温对需要一个独立的热相位或 \(\sigma\) ≳ 0.3 dex。M31 自身的 0.1–0.3 keV ISM 得到支持（Kavanagh+2020，Zhang+2019） |
| R078 R078 | 2026-11-03 14:12 EST2026-11-03 14:12 EST | V22V22 | XMM spectrum in B10's Chandra disk rectangleB10 钱德拉圆盘矩形内的 XMM 谱 | open -> narrowed (convention-dependent)开放 -> 收窄（约定依赖） | Spill-corrected (primary): XMM gives 0.40–\(0.57 \times\) B10 at \(\Gamma\) 2 (0.34–\(0.60 \times\) at \(\Gamma\) 1.7) and a hard excess consistent with zero (0.9–\(1.6\sigma\)); no-spill bracket: 1.55–1.78\(\times\) (1.65–1.89\(\times\)). Mask (b) changes no exclusion. The primary and bracket conventions differ by 3.9–\(4.4 \times\) in H, so the zero point / spill convention, not the aperture, decides. FWC/QPB lineage unverified (mos_back deleted the FWC operands)扣除溢出（主约定）：XMM 在 \(\Gamma\) 2 处给出 B10 的 0.40–0.57\(\times\)（\(\Gamma\) 1.7 处 0.34–0.60\(\times\)），且硬超额与零一致（0.9–\(1.6\sigma\)）；不扣除溢出的括号：1.55–1.78\(\times\)（1.65–1.89\(\times\)）。掩护 (b) 不改变任何排除。主约定与括号约定在 H 中相差 3.9–4.4\(\times\)，因此零点/溢出约定，而非孔径，决定结果。FWC/QPB 血缘未验证（mos_back 删除了 FWC 操作数） |
| R079 R079 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | RT-20RT-20 | E036 halo ~0.76 keV detectionE036 晕 ~0.76 keV 检出 | Provisional (statistical detection holds; attribution open)临时性（统计检出成立；归因开放） | q = 22.426 reproduces; global Bonferroni p = 1.42e-5 (threshold 12.46); dof and saved-start checks pass. Not a q decomposition: the registered decomposition does not split q by field/bin (best-grid session not saved; 11.868 \(\chi^2\) of it unlocated). The MW 0.7 keV halo term (MWhothalo) is switched off in the registered set-up, so the detected SrcApec is the halo fields' ~0.7 keV foreground itself; whether it is Milky Way or M31 CGM is not tested (no fit with MWhothalo free or SWCX)q = 22.426 可复现；全局 Bonferroni p = 1.42e-5（阈值 12.46）；dof 与保存的起始值检验通过。不是 q 分解：已注册的分解不按视场/分箱拆分 q（最佳网格会话未保存；其中 11.868 的 \(\chi^2\) 未定位）。已注册设置中 MW 0.7 keV 晕项（MWhothalo）被关闭，因此检出的 SrcApec 本身就是晕场的 ~0.7 keV 前景；它是银河系还是 M31 CGM 未被检验（无 MWhothalo 自由或 SWCX 拟合） |
| R080 R080 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | RT-22RT-22 | working conclusion 'disk 0.7 keV = unresolved stars + flat foreground'工作结论“圆盘 0.7 keV = 未解析恒星 + 平坦前景” | accepted -> WEAKENED wording被接受 -> 措辞减弱 | Floor \(1.13 \pm 0.08e-7\) vs halo-field \(0.97 \pm 0.12e-7\): difference \(0.16 \pm 0.14e-7\) (\(1.1\sigma\)); allowed model ratios 0.65–1.67 (R050/R051); V7 HF1 over-subtracts in 3/3 outer annuli (2/3 data-rejected). Holds: current fits do not require diffuse M31 gas. Removed: that the stellar decomposition is unique, that the halo level transplants unchanged, and the Milky Way identification. Proposed wording: 'current fits do not require diffuse M31 gas; the emission is consistent with stellar-like emission plus an extended foreground mixture'底板 \(1.13 \pm 0.08e-7\) vs 晕场 \(0.97 \pm 0.12e-7\)：差值 \(0.16 \pm 0.14e-7\)（\(1.1\sigma\)）；允许的模型比例 0.65–1.67（R050/R051）；V7 HF1 在 3/3 外环带过度扣除（2/3 数据被拒）。成立：当前拟合不需要弥散的 M31 气体。移除：恒星分解是唯一的、晕场水平原样移植、以及银河系识别。拟议措辞：“当前拟合不需要弥散的 M31 气体；发射与恒星状发射加上一个扩展的前景混合一致” |
| R081 R081 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | RT-21RT-21 | F1 young/old forecastF1 young/old 预测 | narrowed -> only ~\(1000 \times\) defensible收窄 -> 仅 ~1000\(\times\) 可辩护 | F1's conditional Fisher arithmetic reproduces exactly (\(\sigma\)(R) = 22.83 at R = 100) and its mass bookkeeping is exact, but: continuous \(3\sigma\) threshold 77\(\times\), bootstrap power at \(100 \times\) only 71%, with the floor calibrated \(100 \pm 40\) (threshold 131\(\times\)), with \(\chi^2\) inflation \(100 \pm 129,\) with age-mix pooling 688\(\times\) → only ~\(1000 \times\) is defensibly testable. Dominant systematic: a missing additive floor (13–\(16\sigma\); \(\chi^2\)/dof 31–47 → 1.6–2.7 with it), not the bulge or mass lossF1 的条件性 Fisher 算术精确复现（R = 100 处 \(\sigma\)(R) = 22.83）且其质量记账精确，但：连续 \(3\sigma\) 阈值 77\(\times\)，100\(\times\) 时自举功效仅 71%，在底板校准为 \(100 \pm 40\) 时阈值为 131\(\times\)，在 \(\chi^2\) 通胀为 \(100 \pm 129\) 时，以及在年龄混合聚合时为 688\(\times\) → 仅 ~1000\(\times\) 是可辩护可测试的。主导系统误差：缺失的加性底板（13–\(16\sigma\)；\(\chi^2\)/dof 31–47 → 有底板后 1.6–2.7），而非核球或质量损失 |
| R082 R082 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | Lit L27Lit L27 | intermediate-age (0.1–5 Gyr) emissivity中年龄（0.1–5 Gyr）发射率 | open -> intermediate bin untestable开放 -> 中年龄箱不可测试 | No age-binned per-mass L_X in 0.1–5 Gyr found. Old-field ABs \(1.08 \pm 0.16e28\) and CVs \(2.5 \pm 0.6e27\) erg/s/M☉ (2–10 keV, Warwick 2014); L_X/L_bol saturates \(\approx 1e-3\) below ~100 Myr and declines after Hyades age, not as a power law (Randich 1999, Güdel 2004). DERIVED: 0.3–5 vs > 5 Gyr ratio likely ≲ 10\(\times\)未在 0.1–5 Gyr 中找到按年龄分箱的每质量 L_X。旧场 ABs \(1.08 \pm 0.16e28\) 与 CVs \(2.5 \pm 0.6e27\) erg/s/M☉（2–10 keV，Warwick 2014）；L_X/L_bol 在 ~100 Myr 以下饱和于 \(\approx 1e-3\)，并在 Hyades 年龄后下降，不呈幂律（Randich 1999，Güdel 2004）。推导：0.3–5 与 > 5 Gyr 的比值很可能 ≲ 10\(\times\) |
| R083 R083 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | B1B1 | bulge vs disk vs young emissivity, nine annuli核球 vs 圆盘 vs 年轻发射率，九个环带 | open -> hard disk-like; soft not separable开放 -> 硬带随圆盘；软带不可分离 | Hard 2–7 keV: \(\epsilon\)_disk = 9.51 [7.85, 11.66]e27 erg/s/M☉, bulge unconstrained (−5.45 [−14.7, 3.4]e27): the hard emission follows the disk. Soft 0.5–2 keV: \(\epsilon\)_bulge 0.64 [−0.53, 1.86]e27, \(\epsilon\)_disk 2.11 [1.72, 2.61]e27, poor \(\chi^2\) (46/6) because no additive floor was fitted (R081). Young term: soft 4.6 [2.2, 7.1]e37 erg/s per M☉/yr (Mineo hot ISM 8.3e38: \(\times 18\) lower), hard 1.4 [−0.4, 3.2]e38; strongly covariant硬带 2–7 keV：\(\epsilon\)_disk = 9.51 [7.85, 11.66]e27 erg/s/M☉，核球未约束（−5.45 [−14.7, 3.4]e27）：硬发射随圆盘。软带 0.5–2 keV：\(\epsilon\)_bulge 0.64 [−0.53, 1.86]e27，\(\epsilon\)_disk 2.11 [1.72, 2.61]e27，\(\chi^2\) 较差（46/6），因为未拟合加性底板（R081）。年轻项：软 4.6 [2.2, 7.1]e37 erg/s 每 M☉/yr（Mineo 热 ISM 8.3e38：低 18 倍），硬 1.4 [−0.4, 3.2]e38；强相关 |
| R084 R084 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | XLF1XLF1 | faint-XRB share from the New-ANGELS II XLFs来自 New-ANGELS II XLFs 的暗弱 XRB 份额 | open -> narrowed (mask completeness needed)开放 -> 收窄（需要掩护完备度） | 36-variant sweep: faint-XRB 2–7 keV SB 1.7e-18–5.4e-16 erg/cm\(^2\)/s/arcmin\(^{-2}\), 0.1–17% of the measured total hard; primary \(\Gamma\) 1.6: 4.6–16.5% of total hard and 24–85% of the positive same-depth excess in a016_024–a056_064. A power-law soft extrapolation would reach 0.68–\(2.35 \times\) the measured SrcApec in the inner six annuli, so the soft closure is not usable as is. CHALLENGE: needs a matched per-field mask-completeness / removal-limit product36 变体扫描：暗弱 XRB 2–7 keV SB 1.7e-18–5.4e-16 erg/cm\(^2\)/s/arcmin\(^{-2}\)，占测得总硬量的 0.1–17%；主 \(\Gamma\) 1.6：总硬量的 4.6–16.5% 以及 a016_024–a056_064 中正同深度超额的 24–85%。软带幂律外推会达到测得 SrcApec 的 0.68–2.35 倍（内六个环带），因此软闭合暂不可用。挑战：需要每视场匹配的掩护完备度 / 移除限值产品 |
| R085 R085 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | V23V23 | SP shape and MOS-line refits for the inner hard excess内硬超额的 SP 形状与 MOS 线重拟合 | open -> disfavoured开放 -> 不利 | a016_024: freeing the SP shape lowers \(\chi^2\) by 61.95 but moves H by +\(0.03e-15 \pm 0.29e-15\) (\(0.10\sigma\)); excess stays \(1.078 \pm 0.416e-15\); all 112 added MOS-line norms stay at zero; a032_040 and a072_080 controls agreea016_024：放开 SP 形状使 \(\chi^2\) 降低 61.95，但 H 移动 +\(0.03e-15 \pm 0.29e-15\)（\(0.10\sigma\)）；超额仍为 \(1.078 \pm 0.416e-15\)；全部 112 个新增 MOS 线 norm 保持在零；a032_040 与 a072_080 对照一致 |
| R086 R086 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | V2bV2b | E036 kT 0.70 norm intervalE036 kT 0.70 norm 区间 | refined细化 | Nine-point nuisance-refit norm profile at kT 0.70: \(\chi^2\)_min 1252.494 at norm 5.0e-8 arcmin\(^{-2}\) (dof 749); \(\Delta\)\(\chi^2\) \(\leq 2.706\) interval [3.56, 7.43]e-8 (interpolated; grid [4, 7]e-8). The free fit at 0.70 (1254.885, norm 3.64e-8) was not at the optimum, so the registered q values are conservative (q at 0.70 would be 23.47)kT 0.70 处九点 nuisance-refit norm 剖面：\(\chi^2\)_min 1252.494 处 norm 5.0e-8 arcmin\(^{-2}\)（dof 749）；\(\Delta\)\(\chi^2\) \(\leq 2.706\) 区间 [3.56, 7.43]e-8（插值；网格 [4, 7]e-8）。0.70 处的自由拟合（1254.885，norm 3.64e-8）未在最优处，因此注册的 q 值是保守的（0.70 处 q 将为 23.47） |
| R087 R087 | 2026-11-03 17:39 EST2026-11-03 17:39 EST | Lit L26Lit L26 | M31's own soft (~0.1–0.4 keV) diffuse gasM31 自身的软（~0.1–0.4 keV）弥散气体 | open -> M31 gas well motivated开放 -> M31 气体动机充分 | Bulge gas kT 0.2–0.4 keV (Zhang+2019 RGS 0.20–0.25; Shirey+2001 ~0.35, L \(\approx 2e38\); Bogdán & Gilfanov 2008 0.31–0.36; Li & Wang 2007 ~0.4, \(\geq 2.5\) kpc off-plane); northern disk rings kT 0.10–0.33 keV (Kavanagh+2020), following the arms/10-kpc ring and massive stars, fainter and cooler between rings; the gas lies behind M31's near-side HI (internal N_H 1–3e21). In 2° halo fields the Local Hot Bridge (Qu+2021, flat within 15°) is degenerate with an M31-halo 0.2–0.3 keV term核球气体 kT 0.2–0.4 keV（Zhang+2019 RGS 0.20–0.25；Shirey+2001 ~0.35，L \(\approx 2e38\)；Bogdán & Gilfanov 2008 0.31–0.36；Li & Wang 2007 ~0.4，\(\geq 2.5\) kpc 盘外）；北圆盘环带 kT 0.10–0.33 keV（Kavanagh+2020），沿臂/10-kpc 环和大质量恒星分布，臂间更暗更冷；气体位于 M31 近侧 HI 之后（内部 N_H 1–3e21）。2° 晕场视场中，Local Hot Bridge（Qu+2021，15° 内平坦）与 M31-晕 0.2–0.3 keV 项退化 |
| R088 R088 | 2026-11-03 22:00 EST2026-11-03 22:00 EST | RT-23RT-23 | B1 hard \(\epsilon\)_diskB1 硬 \(\epsilon\)_disk | Provisional -> conditional临时性 -> 有条件 | Bulge/disk template choices move \(\epsilon\)_disk \(\leq 1.7\)%, but the halo zero point gives −4.57 [−13.4, 3.9]e27 and the primary total-XRB correction 5.81 [4.55, 7.37]e27 (2–7 keV): disk-like only under the adopted same-depth zero and not XRB-free核球/圆盘模板选择使 \(\epsilon\)_disk 变动 \(\leq 1.7\)%，但晕零点给出 −4.57 [−13.4, 3.9]e27，主总-XRB 修正给出 5.81 [4.55, 7.37]e27（2–7 keV）：仅在采用的同深度零点下是圆盘状，且不是无 XRB 的 |
| R089 R089 | 2026-11-03 22:00 EST2026-11-03 22:00 EST | RT-24RT-24 | faint-XRB closure of the inner hard excess内硬超额的暗弱 XRB 闭合 | narrowed -> uncalibrated收窄 -> 未校准 | Closure 23.8–84.7% nominal, 15.2–55.4% with a \(\times 2\) fainter removal limit, 37.5–129.5% with \(\times 2\) brighter; on the spill-corrected residual-over-M32 denominator 29.8–87.5%; mostly HMXBs. Faint XRBs stay a credible contributor, not a calibrated closure名义闭合 23.8–84.7%，\(\times 2\) 更暗移除限值下 15.2–55.4%，\(\times 2\) 更亮下 37.5–129.5%；在扣除溢出的残余-over-M32 分母下为 29.8–87.5%；绝大部分为 HMXB。暗弱 XRB 仍是可信的贡献者，而非已校准的闭合 |
| R090 R090 | 2026-11-03 22:00 EST2026-11-03 22:00 EST | Lit L28Lit L28 | per-mass coronal X-ray emission vs age按质量的日冕 X 射线发射 vs 年龄 | open -> expected contrast far below the threshold开放 -> 期望对比度远低于阈值 | DERIVED from Núñez & Agüeros 2016 per-star medians \(\times\) Kroupa IMF: L_X/M (0.5–7 keV) ~7e29 at 35 Myr, ~3e29 at 0.2–0.5 Gyr, ~5e28 at 625 Myr, vs ~1.3e28 erg/s/M☉ for old-field ABs + CVs (Warwick 2014): stellar-only (0–0.3)/(> 5 Gyr) \(\approx 20\)–50\(\times\), (0.3–5)/(> 5) ≲ 10\(\times\); \(\geq\) \(\times 3\) uncertainty; no cluster-integrated L_X/M found从 Núñez & Agüeros 2016 每恒星中位数 \(\times\) Kroupa IMF 推导：L_X/M（0.5–7 keV）~7e29 处 35 Myr，~3e29 处 0.2–0.5 Gyr，~5e28 处 625 Myr，而旧场 ABs + CVs 为 ~1.3e28 erg/s/M☉（Warwick 2014）：仅恒星 (0–0.3)/(> 5 Gyr) \(\approx 20\)–50\(\times\)，(0.3–5)/(> 5) ≲ 10\(\times\)；\(\geq\) \(\times 3\) 不确定度；未找到团簇积分 L_X/M |
| R091 R091 | 2026-11-04 00:59 EST2026-11-04 00:59 EST | RC1RC1 | regression with a free additive floor and per-camera normalisation带自由加性底板与逐相机归一化的回归 | Provisional -> tighter, shape-conditional临时性 -> 更紧、有条件于形状 | Free floor + MOS1/MOS2 scale repair 0.8–2 keV (\(\chi^2\)/dof 1.37) but not 0.5–0.8 (3.08). Hard \(\epsilon\)_disk \(1.28e28 \pm 0.05e28\) erg/s/M☉ after faint-XRB removal (−6%). Diffuse gas: only the HI-shaped term is separable — coef \(3.47 \pm 1.56e-17\) (\(2.2\sigma\)), \(3\sigma\) L_X(0.5–2) < 3.8e37 / 4.4e37 erg/s (inner / outer a = 56′), ~\(6 \times\) below R064; anti-correlated with the floor (−0.75); flat and SFR-shaped gas are degenerate with the floor and young terms. Injection–recovery unbiased for the HI shape. No pn product, so no pn normalisation自由底板 + MOS1/MOS2 尺度修复 0.8–2 keV（\(\chi^2\)/dof 1.37）但不是 0.5–0.8（3.08）。扣除暗弱 XRB 后硬 \(\epsilon\)_disk \(1.28e28 \pm 0.05e28\) erg/s/M☉（−6%）。弥散气体：只有 HI 形状项可分离——系数 \(3.47 \pm 1.56e-17\)（\(2.2\sigma\)），\(3\sigma\) L_X(0.5–2) < 3.8e37 / 4.4e37 erg/s（内 / 外 a = 56′），比 R064 低约 6 倍；与底板反相关（−0.75）；平坦与 SFR 形状的气体与底板及年轻项退化。HI 形状项的注入–恢复无偏差。无 pn 产品，故无 pn 归一化 |
| R092 R092 | 2026-11-04 00:59 EST2026-11-04 00:59 EST | RT-25RT-25 | expected young/old contrast期望的 young/old 对比度 | re-scope withdrawn重新范围已撤回 | Coronal-only contrast is 10–\(35 \times\) on the PHAT formed-mass basis (L28's living-mass divisor \(2.17 \times\) too small), \(\approx 4\)–\(24 \times\) in 2–7 keV. Adding sourced HMXBs (Gilbertson+2022: L_HMXB/SFR = 10^39.68 = 4.8e39) and SNRs (Maoz & Badenes 2010) gives a total (0–0.3 Gyr)/(> 5 Gyr) \(\approx\) \(2200 \times\) for a flat 0–300 Myr SFH, 115–\(8350 \times\) across SFH shapes — at or above the ~\(1000 \times\) threshold. The deciding unknown is the SFH inside the 0–0.3 Gyr bin (activity younger than the ~84 Myr HMXB delay)仅日冕对比度在 PHAT 已形成质量基础上为 10–35\(\times\)（L28 的在世质量除数 2.17\(\times\) 过小），2–7 keV 中 \(\approx 4\)–24\(\times\)。加上来源 HMXBs（Gilbertson+2022：L_HMXB/SFR = 10^39.68 = 4.8e39）与 SNRs（Maoz & Badenes 2010），总计 (0–0.3 Gyr)/(> 5 Gyr) 对于平坦的 0–300 Myr SFH \(\approx 2200\)\(\times\)，跨 SFH 形状为 115–8350\(\times\)——达到或超过 ~1000\(\times\) 阈值。决定性的未知是 0–0.3 Gyr 箱内的 SFH（比 ~84 Myr HMXB 延迟更年轻的活动） |
| R093 R093 | 2026-11-04 00:59 EST2026-11-04 00:59 EST | RT-26RT-26 | halo-field level after the V2b norm optimumV2b norm 最优后的晕场水平 | Provisional -> Questioned (one common level)临时性 -> 受质疑（一个共同水平） | The \(\times 1.37\) V2b step moves floor − halo by only −\(0.90\sigma\), but the level is estimator-dependent: v19 (R027) \(9.66 \pm 1.16e-8\) vs E036 5.0 [3.56, 7.43]e-8, 2.3–\(2.7\sigma\) apart; registered halo levels span \(\times 11.8\) across fits of the same data. Floor − halo = +\(1.64 \pm 1.41e-8\) (\(1.2\sigma\)) with v19, +\(6.30 \pm 1.52e-8\) (\(4.2\sigma\); \(3.4\sigma\) inflated) with E036. Detection strengthened (q \(\geq 23.47,\) p \(\leq 8.3e-6\))\(\times 1.37\) 的 V2b 步使 floor − halo 仅移动 −\(0.90\sigma\)，但水平依赖于估计器：v19（R027）\(9.66 \pm 1.16e-8\) vs E036 5.0 [3.56, 7.43]e-8，相差 2.3–\(2.7\sigma\)；已注册的晕水平在同一数据的拟合间跨度 \(\times 11.8\)。Floor − halo：v19 下为 +\(1.64 \pm 1.41e-8\)（\(1.2\sigma\)），E036 下为 +\(6.30 \pm 1.52e-8\)（\(4.2\sigma\)，通胀后 \(3.4\sigma\)）。检出加强（q \(\geq 23.47\)，p \(\leq 8.3e-6\)） |
| R094 R094 | 2026-11-04 00:59 EST2026-11-04 00:59 EST | Lit L29Lit L29 | MW soft foreground around M31 at 2–15°M31 周围 2–15° 的 MW 软前景 | open -> design inputs开放 -> 设计输入 | Local Hot Bridge (Qu+2021) flat to \(\approx 15\)° (\(\approx 209\) kpc): O VII \(5.6 \pm 1.3,\) O VIII \(2.8 \pm 0.6\) L.U., log T 6.34 (\(\approx 0.19\) keV) → no off-field within 15° removes the 0.2–0.3 keV foreground. eRASS1 DR1 does not cover M31 (l 121° < 180°). Suzaku 0.7 keV field scatter P16–P84 = 2.1–9.5e-4 cm\(^{-6}\) pc (\(\times 4.5\)); no Suzaku field 0.7–15.5° from M31. Galactic mirror field l = 238.83°, b = −21.57° lies in eRASS1 DR1Local Hot Bridge（Qu+2021）至 \(\approx 15\)°（\(\approx 209\) kpc）平坦：O VII \(5.6 \pm 1.3\)，O VIII \(2.8 \pm 0.6\) L.U.，log T 6.34（\(\approx 0.19\) keV）→ 15° 内无离场视场能移除 0.2–0.3 keV 前景。eRASS1 DR1 不覆盖 M31（l 121° < 180°）。Suzaku 0.7 keV 视场散射 P16–P84 = 2.1–9.5e-4 cm\(^{-6}\) pc（\(\times 4.5\)）；距 M31 0.7–15.5° 无 Suzaku 视场。银河系镜像视场 l = 238.83°，b = −21.57° 在 eRASS1 DR1 中 |
| R095 R095 | 2026-11-04 05:01 EST2026-11-04 05:01 EST | V8V8 | clipped-sector extraction pilot, one ObsID, D24 full-ODF v1.7.11剪切扇区提取试点，单 ObsID，D24 完整 ODF v1.7.11 | pilot -> PASS_WITH_RECORDED_OUTCOMES试点 -> 记录了结果的通过 | 19/30 receipt tasks published (12 sector camera spectra in 7 regions, 5 with both cameras, 7 parent controls), 2 failures, 29 skips; 7/10 parent camera pairs cross-version comparable. Corrected 33-ObsID gate: 452 camera tasks, 1821 h one-worker wall time, 197.9 GiB (10% overhead). fast_regions comparison not run (V12)19/30 个收据任务已发布（7 个区域中的 12 个扇区相机谱，5 个两台相机都有，7 个父母控制），2 个失败，29 个跳过；7/10 个父母相机对跨版本可比。修正后的 33-ObsID 门控：452 个相机任务，1821 h 一工人墙钟时间，197.9 GiB（10% 开销）。未运行 fast_regions 比较（V12） |
| R096 R096 | 2026-11-04 05:01 EST2026-11-04 05:01 EST | XLF2XLF2 | per-field removal limit and completeness每视场移除限值与完备度 | closure 11–45%, not the whole excess闭合 11–45%，并非整个超额 | Masks miss no eligible catalogue source (membership efficiency 1.000 in all 465 fields); area-weighted 90% removal completeness 2.3–6.2e-15 erg cm\(^-2\) s\(^-1\) (2–4.5 keV). Completeness weighting cuts the \(\Gamma\) = 1.6 faint-XRB hard level to 0.457–0.531 of the hard cut (aggregate 0.479); inner-six closure 11.3–44.9% (R084 same-depth) and 15.8–40.4% (R085), aggregate 22.8% / 27.1%, denominator-only errors. "HMXB" is 72–90% of the faint hard XRB (\(\approx 7.1e37\) erg/s per M☉/yr below the limit; dispatcher sum)掩护未遗漏任何合格目录源（全部 465 个视场的成员效率 1.000）；面积加权的 90% 移除完备度 2.3–6.2e-15 erg cm\(^-2\) s\(^-1\)（2–4.5 keV）。完备度加权将 \(\Gamma\) = 1.6 暗弱 XRB 硬量削减为硬切口的 0.457–0.531（总计 0.479）；内六个环带闭合 11.3–44.9%（R084 同深度）与 15.8–40.4%（R085），总计 22.8% / 27.1%，分母仅误差。“HMXB”占暗弱硬 XRB 的 72–90%（低于限值以下约 7.1e37 erg/s 每 M☉/yr；调度方求和） |
| R097 R097 | 2026-11-04 05:01 EST2026-11-04 05:01 EST | D1aD1a | PHAT 0–0.3 Gyr split + unresolved young forecast with a free floorPHAT 0–0.3 Gyr 分割 + 带自由底板的未解析年轻预测 | v1 PROVISIONAL, sent back (CORRECTION_01)v1 临时性，退回（CORRECTION_01） | Split (WISE 22 µm 100-Myr proxy \(\times\) Lewis+2015 Table 2 shape): median 1.67e7 M☉ in 0–100 Myr, 2.77e7 in 100–300 Myr. Young template 2.86e37 (0.5–2) / 2.48e37 (2–7 keV) erg/s, SNR-dominated (\(\Gamma\) = 2 placeholder). Fisher forecast (MOS1+MOS2+pn, free floor, camera nuisances): soft forecast detectable (SNR 4.33, min. detectable scale 0.685), hard not (5.21); all-regions grouping beats young-fraction stacks. Dispatcher catch: the unresolved-HMXB term multiplies "SB per annulus SFR" by the PHAT region area (1.91 arcmin\(^{-2}\)) instead of the annulus area, so it is low by ~170–620\(\times\) (script lines 492–509; XLF1 sfr_factor is the annulus total); expected ~1e37 erg/s, script 1.1e35分割（WISE 22 µm 100-Myr 代理 \(\times\) Lewis+2015 Table 2 形状）：0–100 Myr 中位数 1.67e7 M☉，100–300 Myr 中位数 2.77e7。年轻模板 2.86e37（0.5–2）/ 2.48e37（2–7 keV）erg/s，SNR 主导（\(\Gamma\) = 2 占位符）。Fisher 预测（MOS1+MOS2+pn，自由底板，相机 nuisance）：软预测可检出（SNR 4.33，最小可检测尺度 0.685），硬不可（5.21）；全区域分组优于年轻分数堆叠。调度方发现：未解析 HMXB 项将“每环带 SB \(\times\) SFR”乘以 PHAT 区域面积（1.91 arcmin\(^{-2}\)）而非环带面积，因此偏低 ~170–620\(\times\)（脚本第 492–509 行；XLF1 sfr_factor 是环带总量）；期望 ~1e37 erg/s，脚本 1.1e35 |
| R098 R098 | 2026-11-04 05:01 EST2026-11-04 05:01 EST | RT-27RT-27 | red team on RC1's HI-shaped gas limit红队审 RC1 的 HI 形状气体上限 | holds as arithmetic; conditional envelope 2.3–6.2e37 / 2.7–7.1e37 erg/s作为算术成立；条件性区间 2.3–6.2e37 / 2.7–7.1e37 erg/s | RC1's g = \(3.47 \pm 1.56e-17\) and the 3.77e37/4.37e37 limits reproduce exactly (22 nested-fit and injection gates pass). The profile term's significance spans 1.06–\(4.29\sigma\) across floor anchors, annuli, bands and cameras, so it is not a robust detection; with the halo floor fixed at the v19 level (9.66e-8) the limit is 2.35e37/2.72e37, at the E036 minimum (5e-8) 3.72e37/4.32e37; the per-camera fit gives 4.11e37/4.77e37. CHALLENGE accepted: RC1's task-2 HI limit did not use its per-camera rowsRC1 的 g = \(3.47 \pm 1.56e-17\) 与 3.77e37/4.37e37 上限精确复现（22 个嵌套拟合与注入门控全部通过）。该项显著性在底板锚点、环带、波段与相机间跨度为 1.06–\(4.29\sigma\)，故不是稳健检出；若晕底板固定在 v19 水平（9.66e-8），上限为 2.35e37/2.72e37，在 E036 最小值（5e-8）处为 3.72e37/4.32e37；逐相机拟合给出 4.11e37/4.77e37。挑战被接受：RC1 的任务 2 HI 上限未使用其逐相机行 |
| R099 R099 | 2026-11-04 05:01 EST2026-11-04 05:01 EST | RT-28RT-28 | is the ~\(2200 \times\) young/old contrast right for the unresolved emission?~2200\(\times\) 的 young/old 对比度对未解析发射是否正确？ | refuted for the unresolved flat-SFH case: 60–222\(\times\)对未解析平坦 SFH 案例被反驳：60–222\(\times\) | Removing resolved HMXBs (7e35 proxy: unresolved HMXB share 4.2%) gives 60–\(222 \times\) under the two primary removal proxies (39–\(395 \times\) over the normalisation sweep); 2–7 keV 16.5–245\(\times\), below the ~\(1000 \times\) threshold; 0.5–2 keV 112–1680\(\times\), conditional on an unsourced old-soft emissivity and on diffuse SNRs left in the spectra (156 Lee/OSC SNRs crossmatched to 465 spectra)移除已解析 HMXBs（7e35 代理：未解析 HMXB 份额 4.2%）在两个主移除代理下给出 60–222\(\times\)（归一化扫描 39–395\(\times\)）；2–7 keV 16.5–245\(\times\)，低于 ~1000\(\times\) 阈值；0.5–2 keV 112–1680\(\times\)，条件于未来源的旧软发射率及谱中残留的弥散 SNR（465 个谱中交叉匹配了 156 个 Lee/OSC SNR） |
| R100 R100 | 2026-11-04 06:05 EST2026-11-04 06:05 EST | V12b / E044 stage BV12b / E044 阶段 B | fast_regions rc4 vs native v1.7.11 on the V8 clipped-sector pilotV8 剪切扇区试点上的 fast_regions rc4 vs 原生 v1.7.11 | BLOCKED: rc4 cannot extract narrow sectorsBLOCKED：rc4 无法提取窄扇区 | rc4 fast_regions/selection.py:133-135 raises "Unsupported DET shape POLYGON" in 12/12 sector tasks (3 excluded polygons \(\times 110\) vertices); 7 parent tasks compile but were not extracted; no FWC operands kept natively. Independent area diagnostics vs native BACKSCAL +3.27/+1.83/+2.81% (uncertified, HOLD). Native cost 5794 s = 1.61 h per sector task (the brief's 3.7 h was pilot wall time / published count: dispatcher brief error, CHALLENGE upheld). V8b 3 matched annuli = 258 tasks \(\approx 457\) h native commands (1039 h on the pilot allocation). SLCTEXPR header \(\neq\) executed evselect expression in 7/19 native cells (header truncated, command correct)rc4 fast_regions/selection.py:133-135 在 12/12 个扇区任务中抛出 "Unsupported DET shape POLYGON"（3 个排除多边形 \(\times 110\) 个顶点）；7 个父母任务可编译但未提取；未原生保留任何 FWC 操作数。与原生 BACKSCAL 的独立面积诊断 +3.27/+1.83/+2.81%（未认证，HOLD）。原生成本每扇区任务 5794 s = 1.61 h（简报中的 3.7 h 是试点墙钟时间 / 发布计数：调度方简报错误，挑战维持）。V8b 的 3 个匹配环带 = 258 个任务 \(\approx 457\) h 原生命令（试点分配下 1039 h）。7/19 个原生单元中 SLCTEXPR 标头 \(\neq\) 已执行的 evselect 表达式（标头被截断，命令正确） |
| R101 R101 | 2026-11-04 06:05 EST2026-11-04 06:05 EST | D1a CORRECTION_01D1a CORRECTION_01 | HMXB area fix, known-answer gateHMXB 面积修正，已知答案门控 | v2 accepted as a forecast, conditionalv2 作为有条件预测被接受 | HMXB young term 6.060e37 soft / 2.469e37 hard erg/s (dispatcher estimate ~2.4e37 hard); totals 8.893e37 / 4.941e37; known-answer gate a016_024 8.1036e36, relative difference 0. Forecast SNR soft 5.02 (minimum detectable scale 0.585), hard 0.63 (6.80); the all-regions grouping is best. Partly superseded: the photon-vs-energy flux error (R102) makes the soft HMXB ~\(3 \times\) too high and the LMXB reading removes most of itHMXB 年轻项 6.060e37 软 / 2.469e37 硬 erg/s（调度方估计 ~2.4e37 硬）；总计 8.893e37 / 4.941e37；已知答案门控 a016_024 8.1036e36，相对差 0。预测 SNR 软 5.02（最小可检测尺度 0.585），硬 0.63（6.80）；全区域分组最佳。部分被取代：光子-vs-能量通量误差（R102）使软 HMXB 偏高约 3\(\times\)，LMXB 读法移除了其中大部分 |
| R102 R102 | 2026-11-04 06:05 EST2026-11-04 06:05 EST | RT-29RT-29 | faint "HMXB" vs LMXB relabelling in XLF1/XLF2XLF1/XLF2 中暗弱 "HMXB" vs LMXB 重贴标签 | NOT ESTABLISHED; plus a units error未确立；加上一个单位错误 | In the PI's New-ANGELS II fits LMXB-only has lower C than HMXB-only in every disk region (All 60.53/99 vs 80.41/100; \(\Delta\)C Disk 9.71, InnerArm 12.38, Ring 8.39, InterArm 1.17; nuisance sets differ). Under the LMXB-only reading the inner-six closure is 3.5% / 4.1% (additive 22.8 / 27.1%). XLF1 band_int integrates E^(−\(\Gamma\)) (photons) where an energy fraction needs E^(1−\(\Gamma\)): soft fractions \(\times 3.0\) too high, hard \(\times 0.81\); the energy fix alone raises the hard closure by +19.6 points (27.3 / 32.4%)PI 的 New-ANGELS II 拟合中，LMXB-only 在每个圆盘区域的 C 都低于 HMXB-only（全部 60.53/99 vs 80.41/100；\(\Delta\)C 圆盘 9.71，内臂 12.38，环带 8.39，星间 1.17；nuisance 集合不同）。在仅 LMXB 读法下，内六个环带闭合为 3.5% / 4.1%（加性 22.8 / 27.1%）。XLF1 band_int 对 E^(−\(\Gamma\))（光子）积分，而能量分数需要 E^(1−\(\Gamma\))：软分数偏高 \(\times 3.0\)，硬分数 \(\times 0.81\)；仅能量修正就使硬闭合提高 +19.6 个百分点（27.3 / 32.4%） |
| R103 R103 | 2026-11-04 06:05 EST2026-11-04 06:05 EST | Lit L30Lit L30 | FG1 mirror and matched-latitude fieldsFG1 镜像与匹配纬度视场 | field list ready视场列表已就绪 | The brief's two criteria define disjoint sets (mirror is 105.45° from M31: CHALLENGE upheld, dispatcher brief error). XMM set A 63 pointings 3443.9 ks; set B 13 pointings 809.9 ks (deepest ESO 364-29 308.7 + 152.4 ks). Suzaku OFF-FIELD3 509044010 (l 91.74°, b −24.10°, 27.2° from M31, 97.3 ks): kT \(0.799 \pm 0.121\) keV, EM (\(0.59 \pm 0.24\))e-3 cm\(^{-6}\) pc. eRASS1 tile ERASS1_094123 downloaded, not analysed简报的两个标准定义了不相交的集合（镜像视场距 M31 105.45°：挑战维持，调度方简报错误）。XMM 集 A 63 个点 3443.9 ks；集 B 13 个点 809.9 ks（最深 ESO 364-29 308.7 + 152.4 ks）。Suzaku OFF-FIELD3 509044010（l 91.74°，b −24.10°，距 M31 27.2°，97.3 ks）：kT \(0.799 \pm 0.121\) keV，EM (\(0.59 \pm 0.24\))e-3 cm\(^{-6}\) pc。eRASS1 瓦片 ERASS1_094123 已下载，未分析 |
| R104 R104 | 2026-11-04 06:05 EST2026-11-04 06:05 EST | Expert research 3专家研究 3 | (astra-dawson high)（astra-dawson 高） | criteria not met; 8 tasks proposed标准未满足；提出了 8 个任务 | A1 V7 intercept double subtraction (refit C should use SrcApec alone: HF1 −\(1.32\sigma\) not −\(4.62\sigma\), HF2 0) — accepted, dispatcher brief wording error, CORRECTION_03 queued for V7. A2 profile convergence (V2b 0.70 point below the grid minimum; kT crossings 0.673–0.859) — accepted, RT-31 launched. A3 one photon-ledger/selection contract — accepted in part: folded into XLF3 (one absorption, energy flux) rather than a new job. A4 X1 → D1 as a joint measurement — accepted, D1 design after XLF3. A5 one-estimator halo level, then FG1 — accepted, FG1 designed from L30 after RT-31. A6 V12b → V8b — accepted (V12b blocked; re-gated). A7 G1 + DEM1 pilot — accepted, G1 after V1b. A8 V5 and record repairs — deferred (low priority, D28). Unasked question to the PI: measure "what is left after a stellar model" or "how much hot gas, including gas co-spatial with stars/SF, the data can exclude"A1 V7 截距双重扣除（重拟合 C 应单独使用 SrcApec：HF1 −\(1.32\sigma\) 而非 −\(4.62\sigma\)，HF2 0）——接受，调度方简报措辞错误，CORRECTION_03 已排入 V7 队列。A2 剖面收敛（V2b 0.70 点低于网格最小值；kT 交叉点 0.673–0.859）——接受，RT-31 已启动。A3 一份光子账簿/选择契约——部分接受：并入 XLF3（一次吸收，能量通量），而非新任务。A4 X1 → D1 作为联合测量——接受，D1 设计在 XLF3 之后。A5 单估计器晕水平，然后 FG1——接受，FG1 在 RT-31 之后从 L30 设计。A6 V12b → V8b——接受（V12b 阻塞；重新门控）。A7 G1 + DEM1 试点——接受，V1b 之后做 G1。A8 V5 与记录修复——推迟（低优先级，D28）。向 PI 的未提问题：测量“恒星模型之后还剩什么”或“包括与恒星/SF 共空间的气体在内，数据能排除多少热气体” |
| R105 R105 | 2026-11-04 08:45 EST2026-11-04 08:45 EST | XLF3XLF3 | energy-flux band integral fix, two readings能量通量带积分修正，两种读法 | corrected; both readings delivered已修正；两种读法均已交付 | Inner-six hard closure additive 27.26% (R084 def.) / 32.36% (R085), LMXB-only 4.22% / 5.01% (as recorded 22.80 / 27.07%). XLF1 soft closure: as recorded 0.69–2.35 of the soft SB (faint XRBs exceeding all soft emission in 5/6 annuli — a sanity failure nobody flagged), energy additive 0.18–0.61, LMXB-only 0.04–0.12. D1a soft HMXB intrinsic 1.588e37 erg/s (absorbed 9.99e36) additive, 0 LMXB-only内六个环带硬闭合加性 27.26%（R084 定义）/ 32.36%（R085），仅 LMXB 4.22% / 5.01%（记录为 22.80 / 27.07%）。XLF1 软闭合：记录为软 SB 的 0.69–2.35（暗弱 XRB 在 5/6 个环带中超过全部软发射——一个未被标记的健全性失败），能量加性 0.18–0.61，仅 LMXB 0.04–0.12。D1a 软 HMXB 内在 1.588e37 erg/s（吸收后 9.99e36）加性，0 仅 LMXB |
| R106 R106 | 2026-11-04 08:45 EST2026-11-04 08:45 EST | RT-30RT-30 | SNR term of the D1a young templateD1a 年轻模板的 SNR 项 | hard SNR term refuted; soft kept, conditional硬 SNR 项被反驳；软项保留，有条件 | With sourced thermal SNR spectra (kT 0.24 keV) the hard SNR term is 2.85e34–2.04e36 erg/s, not 2.208e37; the soft term stays 4.24–4.44e37 erg/s with forecast soft SNR 5.22–5.25 (4.80–5.15 over lifetime and diffuse-fraction sensitivities); baseline tables byte-identical to D1a v2. Rests on a 30-kyr lifetime and a diffuse fraction with no M31 source采用来源的热 SNR 谱（kT 0.24 keV），硬 SNR 项为 2.85e34–2.04e36 erg/s，而非 2.208e37；软项仍为 4.24–4.44e37 erg/s，预测软 SNR 5.22–5.25（生命周期与弥散分数敏感度下 4.80–5.15）；基准表与 D1a v2 字节相同。建立在 30-kyr 生命周期与无 M31 来源的弥散分数之上 |
| R107 R107 | 2026-11-04 08:45 EST2026-11-04 08:45 EST | Lit L31Lit L31 | M31/Local Group SNR X-ray population and spectraM31/本星团 SNR X 射线族群与谱 | sourced spectral template来源的谱模板 | Working SNR template kT \(\approx 0.55\) keV (0.2–0.8), N_H \(\approx 1.7e21\); keeping D1a's total 4.441e37 gives soft/hard 3.70e37 / 7.40e36 (hard proxy 16.7%, range 4.8–25.9%). No completeness-corrected SNR luminosity density published. Dispatcher could not reproduce the 16.7% central hard share: an unabsorbed kT 0.55 keV continuum gives ~5% above 2 keV (\(\approx L\)31's lower bound); RT-30's kT 0.24 gives ~0.1%工作 SNR 模板 kT \(\approx 0.55\) keV（0.2–0.8），N_H \(\approx 1.7e21\)；保留 D1a 的总计 4.441e37 得到软/硬 3.70e37 / 7.40e36（硬代理 16.7%，范围 4.8–25.9%）。未发布经完备度修正的 SNR 光度密度。调度方无法复现 16.7% 的中心硬份额：未吸收的 kT 0.55 keV 连续谱在 2 keV 以上给出约 5%（\(\approx L\)31 的下限）；RT-30 的 kT 0.24 给出 ~0.1% |
| R108 R108 | 2026-11-04 08:45 EST2026-11-04 08:45 EST | Lit L32Lit L32 | M31 HMXB census and class shareM31 HMXB 普查与类别份额 | narrowed: additive is physical, LMXB-dominated a conditional prior收窄：加性是物理的，LMXB-主导是条件性先验 | No direct M31 class-resolved XLF below 1e36 erg/s. Published-XLF extrapolation gives HMXB 25.3–47.3% and LMXB 52.7–74.7% of the 16′–64′ 2–7 keV light below the removal limit (LMXB 61.0–78.3% to a 1e35 floor)无直接的 M31 类别分辨 XLF 低于 1e36 erg/s。已发布 XLF 外推给出 HMXB 25.3–47.3% 与 LMXB 52.7–74.7% 的 16′–64′ 2–7 keV 光（低于移除限值，LMXB 到 1e35 地板为 61.0–78.3%） |
| R109 R109 | 2026-11-04 09:40 EST2026-11-04 09:40 EST | X1X1 | PHAT-footprint X-ray region table — BLOCKEDPHAT 足迹 X 射线区域表 — BLOCKED | counts measured; exposure and QPB not usable; F1 forecast low by ~\(30 \times\) in exposure-area计数已测；曝光与 QPB 不可用；F1 预测在曝光-面积上偏低约 30\(\times\) | 1,410,808 rows (826 regions \(\times 427\) event files \(\times 4\) bands), 2,240,760 counts; literal event-count gate exact (9/15/14/26) and full-box area to 6.75e-9. Exposure: S1's masked exposure maps are not made from the same GTIs as S1's events — dispatcher check on 0109270301 MOS1: event ONTIME01 26536 s, run6 spectra EXPOSURE 26287–26308 s, S1 map EXPOSURE 11514 s (rpc/4.background, 2025-11-29); X1's audit finds ONTIME/map 0.05–99 over 390 maps, 36 maps missing. Hard QPB not available (S1 carries 0.5–2 keV QPB only). Measured counts / F1 forecast 66.6\(\times\) (0.5–0.8), 124.9\(\times\) (0.8–2), 527.3\(\times\) (2–7, QPB/SP not in F1); F1's coverage applies frac \(\times\) mean_exp after zero-filled samples are already in mean_exp (X1/F1 exposure-area ratio 32\(\times\))1,410,808 行（826 个区域 \(\times 427\) 个事件文件 \(\times 4\) 个波段），2,240,760 个计数；字面事件计数门控精确（9/15/14/26）且整盒面积到 6.75e-9。曝光：S1 的掩护曝光图不是由 S1 的事件的同一 GTI 制作——调度方检查 0109270301 MOS1：事件 ONTIME01 26536 s，run6 谱 EXPOSURE 26287–26308 s，S1 图 EXPOSURE 11514 s（rpc/4.background，2025-11-29）；X1 审计发现 390 张图中 ONTIME/map 为 0.05–99，36 张图缺失。不可用硬 QPB（S1 仅携带 0.5–2 keV QPB）。测得计数 / F1 预测：66.6\(\times\)（0.5–0.8），124.9\(\times\)（0.8–2），527.3\(\times\)（2–7，QPB/SP 不在 F1 中）；F1 的覆盖在零填充样本已进入 mean_exp 之后应用 frac \(\times\) mean_exp（X1/F1 曝光-面积比 32\(\times\)） |
| R110 R110 | 2026-11-04 09:40 EST2026-11-04 09:40 EST | RT-31RT-31 | halo-field 0.7 keV profile convergence晕场 0.7 keV 剖面收敛 | holds and strengthens; interval open成立并加强；区间开放 | Six warm restarts on LTD: q = 34.84 (local p \(\times 13\) = 2.3e-8) with best alternative \(\chi^2\) 1241.12 at kT 0.79 keV, norm 4.85e-8 arcmin\(^{-2}\); perturbed nulls do not weaken it. Not converged formally within 15 min per fit; the 90% norm interval at 0.79 keV and the right-hand kT crossing remain open → resumed with 3 approved fitsLTD 上六次热重启：q = 34.84（局部 p \(\times 13\) = 2.3e-8），最佳替代 \(\chi^2\) 1241.12 处 kT 0.79 keV，norm 4.85e-8 arcmin\(^{-2}\)；受扰零假设不削弱它。在每拟合 15 min 内未正式收敛；0.79 keV 处的 90% norm 区间与右侧 kT 交叉点仍开放 → 用 3 个已批准拟合恢复 |
| R111 R111 | 2026-11-04 09:40 EST2026-11-04 09:40 EST | Lit L33Lit L33 | SNR X-ray lifetimes and unresolved shareSNR X 射线寿命与未解析份额 | soft term lowered, forecast weakened软项下调，预测减弱 | Published X-ray-bright SNR lifetimes 13–25 kyr (Maoz & Badenes 2010 Sedov 12.6–22.5 kyr; Leahy & Filipović 2022 X-ray age spans 24.6 kyr SMC, 25.2 kyr LMC, 13–14 kyr Galaxy), not 30 kyr; unresolved share of the soft SNR term 0.76–0.84; together the soft SNR term falls 1.2–\(2.0 \times\) and the forecast soft SNR to 3.2–\(3.5\sigma\) at 20 kyr. SNR luminosity below 1e35 erg/s: 1.7% measured (Sasaki+2018 field), \(\leq 23\)% extrapolated已发表 X 射线亮 SNR 寿命 13–25 kyr（Maoz & Badenes 2010 Sedov 12.6–22.5 kyr；Leahy & Filipović 2022 X 射线年龄跨度 24.6 kyr SMC，25.2 kyr LMC，13–14 kyr 银河系），而非 30 kyr；软 SNR 项的未解析份额 0.76–0.84；两者共同使软 SNR 项下降 1.2–2.0\(\times\)，预测软 SNR 在 20 kyr 处降至 3.2–\(3.5\sigma\)。SNR 光度低于 1e35 erg/s：1.7% 测得（Sasaki+2018 视场），\(\leq 23\)% 外推 |
| R112 R112 | 2026-11-04 12:16 EST2026-11-04 12:16 EST | RT-31 RESUME_01RT-31 RESUME_01 | halo-field kT crossing and norm interval晕场 kT 交叉点与 norm 区间 | holds; interval provisional成立；区间临时性 | Warm 0.82 keV minimum \(\chi^2\) 1246.74; right-hand \(\Delta\)\(\chi^2\) = 2.706 crossing at kT 0.8044 keV; q unchanged 34.84 (best 1241.12 at 0.79 keV). Norm at 0.79 keV: 90% interval [3.15, 5.82]e-8 arcmin\(^{-2}\), upper crossing measured, lower quadratic-extrapolated (the 3.3e-8 probe has \(\Delta\)\(\chi^2\) 2.26). All three fits hit the 900 s cap热 0.82 keV 最小 \(\chi^2\) 1246.74；右侧 \(\Delta\)\(\chi^2\) = 2.706 交叉点在 kT 0.8044 keV；q 保持 34.84（最佳 1241.12 处 0.79 keV）。0.79 keV 处 norm：90% 区间 [3.15, 5.82]e-8 arcmin\(^{-2}\)，上交叉点已测，下交叉点二次外推（3.3e-8 探测点 \(\Delta\)\(\chi^2\) 为 2.26）。三个拟合都达到了 900 s 上限 |
| R113 R113 | 2026-11-04 12:16 EST2026-11-04 12:16 EST | RT-32RT-32 | empirical old template vs explicit faint LMXBs经验 old 模板 vs 显式暗弱 LMXB | M32 double count excluded; F1-mask overlap undeterminedM32 双重计数已排除；F1-掩护重叠未确定 | The M32-equivalent removal limit (8.4–8.8e33 erg/s, 0.5–8 keV) is fainter than XLF3's 1e34 erg/s lower limit, so the M32 emissivity and the explicit LMXB term do not overlap. For the masks actually used on M31 (run6/S1 policy) the overlap is 0–79.9% depending on the uncalibrated removal depth; the old term's maximum hard share is 72.4% (hard cut) or 56.8% (completeness)M32 等价移除限值（8.4–8.8e33 erg/s，0.5–8 keV）比 XLF3 的 1e34 erg/s 下限更暗，因此 M32 发射率与显式 LMXB 项不重叠。对 M31 实际使用的掩护（run6/S1 策略），重叠取决于未校准的移除深度，为 0–79.9%；old 项的最大硬份额为 72.4%（硬切口）或 56.8%（完备度） |
| R114 R114 | 2026-11-04 12:16 EST2026-11-04 12:16 EST | Lit L34Lit L34 | PHAT SFH-map systematics in DTD-style regressionsDTD 风格回归中的 PHAT SFH 图系统误差 | D1 error budget on the SFH sideD1 SFH 侧的误差预算 | Nine studies all propagate per-cell random SFH errors (Abelson+2025 as errors-in-variables), none models spatial covariance. From the PHAT table: 0–0.3 Gyr — four isochrone sets agree in pattern (\(\rho\) 0.93), differ \(1.43 \times\) in amplitude, 46% aggregate random error; > 7.9 Gyr — 1.32\(\times\), 4.9%; 2–6 Gyr bins differ 4–\(18 \times\) with uncorrelated patterns and up to 73% empty cells; cell errors are common-mode (quadrature averaging understates the SFH error 14–22\(\times\)). Survey masses BASTI 5.53e10, MIST 5.35e10, PARSEC 4.89e10, Padova 4.98e10 M☉ reproduce Williams+2017 and F1 to < 0.1%九项研究都传播每单元随机 SFH 误差（Abelson+2025 作为误差在变量中）， none 模型空间协方差。来自 PHAT 表：0–0.3 Gyr——四组同龄线在模式上一致（\(\rho\) 0.93），幅度相差 1.43\(\times\)，聚合随机误差 46%；> 7.9 Gyr——1.32\(\times\)，4.9%；2–6 Gyr 箱相差 4–18\(\times\)，模式不相关且多达 73% 空单元；单元误差为共同模式（方根平均低估了 SFH 误差 14–22\(\times\)）。调查质量 BASTI 5.53e10，MIST 5.35e10，PARSEC 4.89e10，Padova 4.98e10 M☉ 与 Williams+2017 和 F1 的差异 < 0.1% |
| R115 R115 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | V7 / E040V7 / E040 | outer 8' annuli with hot-foreground variants外 8' 环带，带热前景变体 | DONE after CORRECTION_03 — undecided (E040 rule v2)CORRECTION_03 后完成——未决（E040 规则 v2） | Intercept from SrcApec alone: HF1 C = −1.318e−\(16 \pm 9.99\)e−17 (z −1.32; v2 table −\(4.62\sigma\)), HF2 C = 0 (all SrcApec at zero; v2 −\(5.03\sigma\)), HF3 C = −1.418e−16. AN obs/pred 1.000/1.022/1.186 at 68/76/84′; HF1 SrcApec 0.288/0/0 of AN. The v19 MWhotHalo level (0.943 s0) can absorb the outer rise. Caveats: AN has 3 more dof than HF1 (not a same-nuisance \(\Delta\)\(\chi^2\)); SP and MOS-line nuisances at their fences仅 SrcApec 的截距：HF1 C = −1.318e−\(16 \pm 9.99\)e−17（z −1.32；v2 表 −\(4.62\sigma\)），HF2 C = 0（全部 SrcApec 在零；v2 −\(5.03\sigma\)），HF3 C = −1.418e−16。AN obs/pred 在 68/76/84′ 处为 1.000/1.022/1.186；HF1 SrcApec 占 AN 的 0.288/0/0。v19 MWhotHalo 水平（0.943 s0）可以吸收外层上升。警告：AN 比 HF1 多 3 个 dof（不是同 nuisance 的 \(\Delta\)\(\chi^2\)）；SP 与 MOS 线 nuisances 在其栅栏上 |
| R116 R116 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | V1b / E035 stage 2V1b / E035 阶段 2 | remaining 15 profiles (CORRECTION_03)剩余 15 个剖面（CORRECTION_03） | contradicted → N_H 0.0585 preferred, profiles pending被否定 → 偏好 N_H 0.0585，剖面待定 | All 8 annuli prefer N_H 0.0585e22: \(\Delta\)\(\chi^2\) −27.8 to −218.6 at fixed dof; an N_H 0.03 refit from the new minimum recovers only 8.9–37.7. MWhalo norm +7.5–43.1%, SrcApec −1.1 to +25.4%; R007 obs/pred profile changes \(\leq\) +0.32 (a080_088). First profile (a024_032 SrcApec.norm, 9/9 points, all nuisances refitted): \(\Delta\)\(\chi^2\)=2.706 interval 2.1549–2.5324e-7, \(2.45 \times\) the conditional \(\sigma\). 15/16 profiles stopped at the 3 LTD-day gate (1.076 h/point; 6.46 LTD-days projected)全部 8 个环带偏好 N_H 0.0585e22：固定 dof 下 \(\Delta\)\(\chi^2\) −27.8 到 −218.6；从新最小值重拟合 N_H 0.03 仅恢复 8.9–37.7。MWhalo norm +7.5–43.1%，SrcApec −1.1 到 +25.4%；R007 obs/pred 剖面变化 \(\leq\) +0.32（a080_088）。第一个剖面（a024_032 SrcApec.norm，9/9 个点，全部 nuisances 重拟合）：\(\Delta\)\(\chi^2\)=2.706 区间 2.1549–2.5324e-7，是条件性 \(\sigma\) 的 2.45 倍。15/16 个剖面在 3 LTD 日门控处停止（1.076 h/点；预测 6.46 LTD 天） |
| R117 R117 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | RT-34RT-34 | F1 forecast without double coverageF1 预测，无双重覆盖 | holds → shifted成立 → 转移 | With F1's second coverage factor removed and run6 EXPOSURE, the minimum detectable young/old per-mass contrast is \(16.43 \times\) in 0.5–2 keV (\(33.84 \times\) with a floor nuisance) and \(659.78 \times\) in 2–7 keV (1072\(\times\)). R073/R081's blanket "\(\geq 100\)–1000\(\times\)" is false for the soft band; the expected unresolved contrast 60–222\(\times\) (RT-28, R099) is testable in soft, not in hard移除 F1 的第二个覆盖因子并使用 run6 EXPOSURE 后，最小可检测每质量 young/old 对比度为 0.5–2 keV 中 16.43\(\times\)（带底板 nuisance 时 33.84\(\times\)）与 2–7 keV 中 659.78\(\times\)（1072\(\times\)）。R073/R081 的笼统 "\(\geq 100\)–1000\(\times\)" 对软带是错误的；期望的未解析对比度 60–222\(\times\)（RT-28，R099）在软带可测试，在硬带不可 |
| R118 R118 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | X1bX1b | matched exposure maps and per-region QPB for the 826 PHAT regions826 个 PHAT 区域的匹配曝光图与每区域 QPB | BLOCKED → usable (soft), upper limit (hard)BLOCKED → 可用（软带），上限（硬带） | 426 exposure maps rebuilt on the event GTIs (MOS1 gate ratio 0.9977, pn 0.988); 1,407,504 rows, 13,015 duplicate sub-pixel memberships removed; run6 QPB for MOS only (62 files, 200 PHA); 622/826 boxes have a net rate. Rate gate b "failed" (1.82–2.95) only by mixing conventions: in one convention the ratios are soft 1.09–1.16 / hard 1.73 (unvignetted), with PHA area\(\times\)time soft 1.07–1.10 / hard 1.35 (#XMMEA_EM vs FLAG==0). 0202230401 has 1,217 zero-exposure counts (grid edge): excluded426 张曝光图在事件 GTI 上重建（MOS1 门控比例 0.9977，pn 0.988）；1,407,504 行，移除了 13,015 个重复子像素成员；run6 QPB 仅为 MOS（62 个文件，200 PHA）；622/826 个框有净速率。速率门控 b "失败"（1.82–2.95）仅因混合约定：在一种约定下比率为软 1.09–1.16 / 硬 1.73（未消光），PHA 面积\(\times\)时间软 1.07–1.10 / 硬 1.35（#XMMEA_EM vs FLAG==0）。0202230401 有 1,217 个零曝光计数（网格边缘）：已排除 |
| R119 R119 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | RT-33RT-33 | impact of the mismatched exposure maps不匹配曝光图的影响 | open → scoped开放 → 范围界定 | Legacy 4.background maps disagree with their events in 10/10 checked cases (factor 0.0524–99.42). Affected: the S1 share package, X1 and F1 (F1 counts \(\times 1.09\)–1.56; X1 median rates \(\times 0.774\) soft, \(\times 0.801\) hard). Unaffected: every spectral-fit luminosity and the M31CGM spectrum-based surface brightness (they use spectrum EXPOSURE and BACKSCAL, not the maps)传统 4.background 图与其事件在全部 10/10 检查的案例中不一致（因子 0.0524–99.42）。受影响：S1 共享包、X1 与 F1（F1 计数 \(\times 1.09\)–1.56；X1 中位速率软 \(\times 0.774\)，硬 \(\times 0.801\)）。不受影响：每个谱拟合光度与基于谱的 M31CGM 表面亮度（它们使用谱 EXPOSURE 与 BACKSCAL，而非这些图） |
| R120 R120 | 2026-11-04 15:47 EST2026-11-04 15:47 EST | FG1 v0FG1 v0 | dispatcher arithmetic, Suzaku off-field EM vs our halo-field norm调度方算术，Suzaku 离场 EM vs 我们的晕场 norm | new, provisional新，临时性 | Suzaku OFF-FIELD3 EM 0.59e−3 cm\(^{-6}\) pc → APEC norm \(\approx 2.0778\)e\(^{-4}\) EM per arcmin\(^{-2}\) \(\approx 1.23\)e-7 arcmin\(^{-2}\) (\(\pm 0.5\)e-7). Our halo-field 0.79 keV norm 3.2–5.8e-8 (R112) is 0.28–\(0.79 \times\) that. Not yet checked: EM definition (n_e n_H vs n_e\(^2\)) and abundance table of the source |
| R121 R121 | 2026-11-04 19:18 EST2026-11-04 19:18 EST | D1D1 | age-resolved emissivity Poisson regression over the PHAT regionsPHAT 区域上的按年龄分辨发射率泊松回归 | Questioned -> formal result only; physical calibration failed受质疑 -> 仅有形式结果；物理校准失败 | Soft (0.5–2 keV) \(\epsilon_{G1}\) (0–0.3 Gyr) 2.30–3.20e29 erg s\(^{-1}\) per formed M☉ across Padova/BASTI/MIST (Padova 2.75e29, 90 % 1.80–3.71e29); \(\epsilon_{G2}\)–G4 = 0 with 90 % uppers \(\leq 0.76e27\) (G4 0.14–0.44e27), except PARSEC G2 2.7e28; hard all zero (G1 < 1.3–1.4e29). Quality fails: soft QPB scales 2.70/2.73 vs prior \(1 \pm 0.10\) (\(17\sigma\)), deviance/dof 3.59 soft / 3.17 hard, 11/32 soft file floors negative. \(\epsilon_{G4}\) sits \(\geq\) \(7 \times\) below the M32 reference. Dispatcher reading: (i) brief error — the soft band summed 1.4–2.0 keV, the MOS line band excluded by project convention; (ii) per-file floors trade with QPB and can absorb smooth old light; (iii) possible double vignetting. CHALLENGE (certify background/response; M31 H I tracer) accepted软（0.5–2 keV）\(\epsilon_{G1}\)（0–0.3 Gyr）在 Padova/BASTI/MIST 上为 2.30–3.20e29 erg s\(^{-1}\) 每已形成 M☉（Padova 2.75e29，90% 1.80–3.71e29）；\(\epsilon_{G2}\)–G4 = 0，90% 上限 \(\leq 0.76e27\)（G4 0.14–0.44e27），PARSEC G2 2.7e28 除外；硬带全为零（G1 < 1.3–1.4e29）。质量不合格：软 QPB 尺度 2.70/2.73 而先前为 \(1 \pm 0.10\)（\(17\sigma\)），deviance/dof 软 3.59 / 硬 3.17，32 个软文件中 11 个底板为负。\(\epsilon_{G4}\) 至少低于 M32 参考值 7 倍。调度方解读：(i) 简报错误——软波段求和了 1.4–2.0 keV，即项目约定排除的 MOS 线波段；(ii) 每文件底板与 QPB 此消彼长，能吸收平滑的旧星光；(iii) 可能存在双重消光。挑战被接受（认证背景/响应；M31 H I 示踪物） |
| R122 R122 | 2026-11-04 19:18 EST2026-11-04 19:18 EST | RT-35RT-35 | is the outer rise wholly MW foreground?外层上升是否完全是银河系前景？ | undecided -> undetermined, power-limited未决 -> 无法判定，受功效限制 | The foreground increment needed to remove the outer rise is sky-flat: −2.202/−2.202/−1.842e−16 erg s\(^{-1}\) cm\(^{-2}\) arcmin\(^{-2}\) at 68/76/84′, constant −2.108e−16 (0.68 s0), \(\chi^2\) 2.44/2; slope \(\Delta\)\(\chi^2\) 1.74/1. The HF1 level is the inverse-variance mean of the 20 v19 off-M31 fields (+\(0.21\sigma\) above their median), not an extrapolation. But the field scatter (norm 5.7e−8) is 2–\(4 \times\) the whole RT-31 halo range (3.2–5.8e−8): \(2\sigma\) halo limit \(\approx 1.1\)e−7, so no per-annulus test at this precision can split halo from foreground. The rise is in obs/pred (1.000/1.022/1.186), not in absolute SrcApec SB (3.09/2.20/1.84e−16), and grows with N_H 0.0585 (+0.03/+0.07/+0.32). Caveats: two of three increments are boundary values (HF1 SrcApec = 0), conditional errors only移除外层上升所需的前景增量在天空上是平坦的：68/76/84′ 处为 −2.202/−2.202/−1.842e−16 erg s\(^{-1}\) cm\(^{-2}\) arcmin\(^{-2}\)，常数 −2.108e−16（0.68 s0），\(\chi^2\) 2.44/2；斜率 \(\Delta\)\(\chi^2\) 1.74/1。HF1 水平是 20 个 v19 离 M31 视场的逆方差平均（比其中位数高 +\(0.21\sigma\)），不是外推。但视场散度（norm 5.7e−8）是整个 RT-31 晕范围（3.2–5.8e−8）的 2–4 倍：\(2\sigma\) 晕上限 \(\approx 1.1\)e−7，因此在此精度下任何逐环带检验都无法区分晕与前景。上升在 obs/pred（1.000/1.022/1.186）中，而不在绝对 SrcApec SB（3.09/2.20/1.84e−16）中，且随 N_H 0.0585 增长（+0.03/+0.07/+0.32）。警告：三个增量中有两个是边界值（HF1 SrcApec = 0），且只有条件性误差 |
| R123 R123 | 2026-11-04 20:11 EST2026-11-04 20:11 EST | D1bD1b | response/exposure convention uncertified响应/曝光约定未认证 | uncertified -> certified with a normalization systematic未认证 -> 已认证，带一个归一化系统误差 | Method (b), X1b unvignetted \(\Omega\)t \(\times\) run6 annulus ARF, over-predicts the fitted sky-model rate by 7.08 / 11.47 / 11.23 % in three 0109270301 MOS1 annuli, identical in all four bands and equal to X1b \(\Omega\)t / (PHA BACKSCAL area \(\times\) EXPOSURE) (2,847,479 / 2,659,120 = 1.0708). Folding agrees with Sherpa to 1e−15. So no energy-dependent response error and no double vignetting; a support/area normalization between X1b and run6 masks. Dispatcher's 5 % gate sat below the 7–10 % offset X1b had already reported (brief design error). CORRECTION_01: certify (b), \(\pm 11\) % on absolute \(\epsilon\) (not on age ratios), check same-support counts/\(\Omega\)t and one MOS2 annulus, then run D1b方法 (b)，X1b 未消光 \(\Omega\)t \(\times\) run6 环带 ARF，在 0109270301 MOS1 的三个环带中把拟合的天空模型速率高估 7.08 / 11.47 / 11.23 %，四个波段完全相同，且等于 X1b \(\Omega\)t /（PHA BACKSCAL 面积 \(\times\) EXPOSURE）（2,847,479 / 2,659,120 = 1.0708）。折叠与 Sherpa 一致至 1e−15。因此没有能量依赖的响应误差，也没有双重消光；而是 X1b 与 run6 掩护之间的支撑/面积归一化。调度方的 5 % 门控低于 X1b 已报告的 7–10 % 偏移（简报设计错误）。CORRECTION_01：认证 (b)，对绝对 \(\epsilon\) 为 \(\pm 11\) %（对年龄比值不是），检查同支撑计数/\(\Omega\)t 与一个 MOS2 环带，然后运行 D1b |
| R124 R124 | 2026-11-04 20:11 EST2026-11-04 20:11 EST | Lit L35Lit L35 | EPIC soft-band calibration and fitted N_HEPIC 软波段校准与拟合 N_H | new, with a dispatcher correction新，含调度方更正 | Published soft residuals 2–3 % (relative EA), MOS/pn 0–3 %, historical redistribution/contamination changes 5–10 %; a 2–3 % energy-dependent residual moves a fitted N_H by ~(0.9–2.5)e20 cm\(^{-2}\); a uniform residual has no N_H leverage. Recommends MOS \(\geq 0.4\) keV, pn \(\geq 0.3\) keV only with validated reductions, Galactic N_H fixed at the L02 H I/dust value with M31 absorption fitted separately. Dispatcher correction: the report's "V1b needs a 94.5–98.9 % deficit" uses 2.85e21 for a shift of 2.85e20 cm\(^{-2}\); by its own table the V1b shift needs only ~5 %, so soft calibration is not excluded. Its CHALLENGE misreads V1b (0.0585 stays preferred)已发表的软波段残余 2–3 %（相对 EA），MOS/pn 0–3 %，历史上的再分配/污染变化 5–10 %；2–3 % 的能量依赖残余会把拟合的 N_H 移动 ~(0.9–2.5)e20 cm\(^{-2}\)；均匀残余对 N_H 没有杠杆。建议 MOS \(\geq 0.4\) keV，pn \(\geq 0.3\) keV 仅在经过验证的约减下使用，银河系 N_H 固定为 L02 的 H I/尘埃值，M31 吸收单独拟合。调度方更正：报告的“V1b 需要 94.5–98.9 % 的亏损”对 2.85e20 cm\(^{-2}\) 的移动用了 2.85e21；按其自己的表，V1b 的移动只需约 5 %，因此软波段校准并未被排除。其挑战误读了 V1b（0.0585 仍被偏好） |
| D26 D26 | 2026-11-03 00:53 EST2026-11-03 00:53 EST | PI, answering 08 §6 item of revision 28PI，对 08 §6 修订版 28 的条目作答 | LTD's compromised-host rules lifted for this project's executors本项目执行者 LTD 的受损主机规则被解除 | ls, compiling and copying files off LTD are allowed; the usual LTD limits stay (CPU/RAM below 60% to start, /home/rhuang \(\leq 100\) MB, nothing on the root disk, scratch under /data/AI_tmp/, no sudo, other users' files untouched)允许 ls、编译及从 LTD 复制文件；常规 LTD 限制保持（启动前 CPU/RAM 低于 60%，/home/rhuang \(\leq 100\) MB，不写根磁盘，暂放在 /data/AI_tmp/，无 sudo，不碰其他用户文件） | |
| D27 D27 | 2026-11-03 01:04 EST2026-11-03 01:04 EST | PI, after the 0.7 keV summaryPI，0.7 keV 总结之后 | Working conclusion accepted; publication needs spatial and population evidence, including PHAT resolved SFHs接受工作结论；发表需要空间与人口证据，包括 PHAT 已解析星形成历史 | Before a paper: (1) a gas upper limit with a reduced stellar-term uncertainty; (2) bulge vs disk stellar emissivity; (3) young vs old per-mass emissivity; (4) a cross-analysis with PHAT resolved SFHs. An upper limit is an acceptable endpoint在论文之前：(1) 带更小恒星项不确定度的气体上限；(2) 核球 vs 圆盘恒星发射率；(3) 年轻 vs 旧的每质量发射率；(4) 与 PHAT 已解析 SFH 的交叉分析。上限是可接受的终点 | |
| D28 D28 | 2026-11-03 01:11 EST2026-11-03 01:11 EST | PI, autonomous directionPI，自主方向 | Proceed autonomously on D27; joint analysis with M31-\(\tau\)X's data; pn added; V5 kept at low priority在 D27 上自主推进；与 M31-\(\tau\)X 数据联合分析；pn 加入；V5 保持低优先级 | Age-resolved emissivity runs as the main line reading M31-\(\tau\)X's PHAT data read-only; pn joins MOS for region-level analysis (X1/D1); V5 stays low priority on Virgo after C1–C3 fits按年龄分辨发射率作为主线，只读读取 M31-\(\tau\)X 的 PHAT 数据；pn 加入 MOS 用于区域级分析（X1/D1）；V5 在 C1–C3 拟合之后保持低优先级，在 Virgo 上运行 | |
| D29 D29 | 2026-11-03 05:34 EST2026-11-03 05:34 EST | dispatcher, under D28调度方，D28 下 | Use the New-ANGELS II XLFs for the faint-XRB share; add the ~0.3 keV gas question and the log-normal temperature hypothesis用 New-ANGELS II XLFs 做暗弱 XRB 份额；加入 ~0.3 keV 气体问题与对数正态温度假设 | XLF1: integrate the New-ANGELS II LMXB/HMXB XLFs below our point-source removal limit in every accepted annulus and XLF region, in 0.5–2 and 2–7 keV; G1: the ~0.2–0.3 keV disk component — M31 gas or the Milky Way halo — from V1b's corrected-N_H profiles; DEM1: a pilot replacing the 0.2–0.3 + 0.7 keV pair with one log-normal emission-measure distribution on halo fields and two disk annuliXLF1：在每个受理环带与 XLF 区域中，将 New-ANGELS II LMXB/HMXB XLFs 在点源移除限值以下于 0.5–2 与 2–7 keV 积分；G1：~0.2–0.3 keV 圆盘成分——M31 气体还是银河晕——来自 V1b 修正 N_H 的剖面；DEM1：试点，用一个对数正态发射度分布替代 0.2–0.3 + 0.7 keV 双温，在晕场与两个圆盘环带上 | |
| R125 R125 | 2026-11-04 23:39 EST2026-11-04 23:39 EST | D1bD1b | age-resolved emissivity regression, shared sky and per-file SP按年龄分辨发射率回归，共享天空与逐文件 SP | formal -> formal, improved; physical gate still fails形式结果 -> 形式结果，有改进；物理门控仍失败 | Response gate: same-support replay PASS (region 261); MOS2 ratio 1.0948, band-independent (span 4e-16), so no energy-dependent response error and no double vignetting. Soft (0.5-2 keV equivalent, 1e27 erg s\(^{-1}\) per formed M☉): \(\epsilon_{G1}\) 207-239, \(\epsilon_{G4}\) 0.88 (Padova, 90 % 0.19-1.58) / 2.70 (BASTI) / 1.83 (MIST) / 1.74 (PARSEC); old-only fit \(\epsilon\) 3.63 (3.24-4.01) vs M32 \(\approx\) 3.2 at retention 0.6; free per-file floors return \(\epsilon_{G4}\) = 0, confirming R121's absorption. Young/old \(\approx\) 80-270\(\times\) (formal). SFR-shaped soft gas term 1.2-1.5e-16 erg cm\(^{-2}\) s\(^{-1}\) arcmin\(^{-2}\) (90 % > 0). The physical gate still fails: the QPB/SP degeneracy in the brief is not resolved by the certified response响应门控：同支撑重放通过（区域 261）；MOS2 比值 1.0948，与波段无关（跨度 4e-16），因此没有能量依赖的响应误差，也没有双重消光。软带（0.5-2 keV 等效，每已形成 M☉ 1e27 erg s\(^{-1}\)）：\(\epsilon_{G1}\) 207-239，\(\epsilon_{G4}\) 0.88（Padova，90% 0.19-1.58）/ 2.70（BASTI）/ 1.83（MIST）/ 1.74（PARSEC）；仅 old 拟合 \(\epsilon\) 3.63（3.24-4.01），而 M32 \(\approx\) 3.2（保留率 0.6）；自由的逐文件底板给出 \(\epsilon_{G4}\) = 0，确认 R121 的吸收结论。Young/old \(\approx\) 80-270\(\times\)（形式结果）。SFR 形状的软气体项 1.2-1.5e-16 erg cm\(^{-2}\) s\(^{-1}\) arcmin\(^{-2}\)（90% > 0）。物理门控仍失败：简报中的 QPB/SP 退化没有被已认证的响应解决 |
| R126 R126 | 2026-11-04 23:39 EST2026-11-04 23:39 EST | RT-36RT-36 | is the N_H 0.0585 preference calibration or degeneracy?N_H 0.0585 的偏好是校准还是退化？ | preferred -> Questioned: driven by E < 0.5 keV被偏好 -> 受质疑：由 E < 0.5 keV 驱动 | With E < 0.5 keV removed, reduced-global refits prefer N_H 0.03 by \(\Delta\chi^2\) 3.79 (a024_032) and 16.73 (a080_088); freeing the frozen MWhalo abundance leaves 0.03 better (11.53, 16.85). Full all-nuisance E \(\geq\) 0.5 refits did not finish (16-h box). With L35 (R124: a ~5 % soft residual moves N_H by the V1b amount) the 0.0585 preference most likely comes from < 0.5 keV calibration or soft-sky modelling, not from the column. D1c uses 0.03 as primary移除 E < 0.5 keV 后，约化全局重拟合偏好 N_H 0.03，\(\Delta\chi^2\) 3.79（a024_032）与 16.73（a080_088）；放开冻结的 MWhalo 丰度后 0.03 仍更优（11.53、16.85）。完整的全 nuisance E \(\geq\) 0.5 重拟合未完成（16 小时框）。结合 L35（R124：约 5% 的软残余就能把 N_H 移动 V1b 的量），0.0585 的偏好最可能来自 < 0.5 keV 的校准或软天空建模，而非柱密度本身。D1c 以 0.03 为主 |
| R127 R127 | 2026-11-04 23:39 EST2026-11-04 23:39 EST | RT-37RT-37 | X1b QPB per soft sub-band, from detector cornersX1b 分软波段 QPB，来自探测器角区 | holds except 1.4-2.0 keV成立，但 1.4-2.0 keV 除外 | Corner/template (ratio of sums): 0.938 \(\pm\) 0.008 (0.5-0.8 keV), 1.014 \(\pm\) 0.006 (0.8-1.4), 0.883 (2-7), 3.71 (1.4-2.0): 64.8 % of the corner 1.4-2 keV counts are Al K\(\alpha\) (peak 1488 eV) + Si K\(\alpha\) lines missing from the run6 continuum template. D1's soft QPB scale 2.7 is this line band, not soft protons. Corner events carry the out-of-FOV flag, so the selection is stated, not identical. D1c fixes QPB per file from the corners; 1.4-2 keV stays out of the science bands角区/模板（求和比值）：0.938 \(\pm\) 0.008（0.5-0.8 keV）、1.014 \(\pm\) 0.006（0.8-1.4）、0.883（2-7）、3.71（1.4-2.0）：角区 1.4-2 keV 计数的 64.8% 是 run6 连续谱模板中缺失的 Al K\(\alpha\)（峰 1488 eV）+ Si K\(\alpha\) 线。D1 的软 QPB 尺度 2.7 正是这条线带，而非软质子。角区事件带有视场外标记，因此选择是声明的、并非完全相同。D1c 用角区逐文件固定 QPB；1.4-2 keV 仍排除在科学波段之外 |
| R128 R128 | 2026-11-04 23:39 EST2026-11-04 23:39 EST | Lit L36Lit L36 | background treatment in resolved X-ray population regressions已解析 X 射线族群回归中的背景处理 | new新 | Old-population calibrators (0.5-2 keV, per current M☉): M32 (5.4 \(\pm\) 0.7 \(\pm\) 1.6)e27, four dEs (6.0 \(\pm\) 0.5 \(\pm\) 1.8)e27, NGC 3379 (8.2 \(\pm\) 0.8 \(\pm\) 2.5)e27; no published > 1 Gyr coefficient in 0.5-1.4 keV. Practice: sky as one physical spectral model anchored by off-disk controls; QPB, SP and Al/Si lines as separate detector terms; integer-count Poisson; deviance calibrated by simulation; profile the old coefficient jointly with sky/QPB/SP. Known-answer suggestion: reproduce the M32 value from its Chandra ObsIDs旧种群校准体（0.5-2 keV，按当前 M☉）：M32 (5.4 \(\pm\) 0.7 \(\pm\) 1.6)e27、四个 dE 星系 (6.0 \(\pm\) 0.5 \(\pm\) 1.8)e27、NGC 3379 (8.2 \(\pm\) 0.8 \(\pm\) 2.5)e27；在 0.5-1.4 keV 没有已发表的 > 1 Gyr 系数。实践：把天空作为一个由离盘对照锚定的物理谱模型；QPB、SP 与 Al/Si 线作为分开的探测器项；整数计数泊松；用模拟校准 deviance；把 old 系数与 sky/QPB/SP 联合剖面。已知答案建议：用其 Chandra ObsID 复现 M32 值 |
| D30 D30 | 2026-11-05 01:27 EST2026-11-05 01:27 EST | PI, in this dispatcher sessionPI，在本次调度方会话中 | One GTI version for every product: rpc/2.filter (3B.background), never 4.background (2.1filter)每个产品只用一种 GTI 版本：rpc/2.filter（3B.background），绝不用 4.background（2.1filter） | accepted被接受 | Cause found: the S1 inventory script preferred rpc/4.background (built from 2.1filter events, a different GTI algorithm) while the events come from rpc/2.filter; 395/431 S1 exposure maps and 506/852 QPB products are 4.background. Header check: map EXPOSURE / max ONTIME of the 2.filter clean events is 3B.background median 1.0000 (56/395 beyond \(\pm 1\) %, 20 beyond \(\pm 5\) %) against 4.background median 0.893 (264/395 beyond \(\pm 5\) %, range 0.010–19.1); 3B equivalents exist for 395/395 maps and 484/506 QPB products, and 36 rows have no map at all. Decision: every product pairs with the 2.filter events — exposure maps and QPB from 3B.background, or made by eexpmap on the 2.filter clean events (the X1b method) where 3B is missing or its EXPOSURE disagrees with ONTIME beyond 1 %. S1b patches the share package (new masked maps, QPB, MANIFEST, README note); the X1/F1 forecasts stay superseded by X1b/RT-34. Upload and sharing stay the PI’s找到原因：S1 的清单脚本优先选了 rpc/4.background（由 2.1filter 事件构建，GTI 算法不同），而事件来自 rpc/2.filter；395/431 张 S1 曝光图与 506/852 个 QPB 产品是 4.background。标头检查：2.filter 洁净事件的图 EXPOSURE / max ONTIME，3B.background 中位数 1.0000（56/395 超出 \(\pm 1\) %，20 超出 \(\pm 5\) %），而 4.background 中位数 0.893（264/395 超出 \(\pm 5\) %，范围 0.010–19.1）；395/395 张图与 484/506 个 QPB 产品存在 3B 对应物，另有 36 行完全没有图。决定：每个产品都与 2.filter 事件配对——曝光图与 QPB 取自 3B.background，或在 3B 缺失或其 EXPOSURE 与 ONTIME 相差超过 1 % 时用 eexpmap 在 2.filter 洁净事件上生成（X1b 方法）。S1b 修补共享包（新掩蔽图、QPB、MANIFEST、README 说明）；X1/F1 预测仍由 X1b/RT-34 取代。上传与共享仍由 PI 决定 |
| R129 R129 | 2026-11-05 02:23 EST2026-11-05 02:23 EST | D1cD1c | corner-fixed QPB, support-clean sample, merged ages角区固定 QPB、支撑洁净样本、合并年龄 | formal only; brief error仅形式结果；简报有误 | Support rule (mine) removed 605/622 regions: 30,345/590,667 events (5.1 %) on zero-mask/exposure pixels, hard-heavy (13,203 at 2–7 keV), i.e. hole edges; 13 regions/2,926 ct left, 0/24 primary cases pass, ratios undefined. Full-sample row (Padova g2 soft): \(\epsilon_{G1}\) 163, \(\epsilon_{G4}\) 1.99 [1.72, 2.25]e27 per formed M☉, ratio 82, dev/dof 1.74, QPB pull 8.4. CHALLENGE held: RT-37 ratios are shape-only, absolute corner\(\rightarrow\)FOV transfer uncertified \(\rightarrow\) D1d支撑规则（我的）移除了 605/622 个区域：590,667 个事件中有 30,345 个（5.1%）落在零掩护/零曝光像素上，且偏硬（2–7 keV 有 13,203 个），即孔洞边缘；只剩 13 个区域/2,926 计数，24 个主用例中 0 个通过，比值无定义。全样本行（Padova g2 软）：\(\epsilon_{G1}\) 163，\(\epsilon_{G4}\) 1.99 [1.72, 2.25]e27 每已形成 M☉，比值 82，dev/dof 1.74，QPB pull 8.4。质疑维持：RT-37 的比值只有形状意义，绝对的角区\(\rightarrow\)视场传递未认证 \(\rightarrow\) D1d |
| R130 R130 | 2026-11-05 03:14 EST2026-11-05 03:14 EST | D1dD1d | D30 input auditD30 输入审计 | gate worked; CORRECTION_01, resumed门控起作用；CORRECTION_01，已恢复 | 657 dependency hashes pass; 61 upstream mask/object images (make_mask ancestors of the X1b source-hole masks) are rpc/4.background (2.1filter) → remake from 3B (791/791 exist) on Virgo. 79/85 PHA GTI unions differ from the counted events, but PHA EXPOSURE/event LIVETIME median 1.0013, range 0.987–1.014 (accepted as \(\leq\) 1.4 % inside the \(\pm\)11 % response certification); the 25 response-gate failures have PHA/live \(\approx\) 1, so their 1.26–1.9\(\times\) excess is area, not time. X1b map axis proxy 25/32 within 1 %, MOS2 \(\approx\) 0.990, 0505760201 mos1 0.858. Same mask lineage also in S1b → CORRECTION_01 queued for S1b657 个依赖哈希通过；61 张上游掩护/目标图像（X1b 源孔洞掩护的 make_mask 祖先）是 rpc/4.background（2.1filter）→ 在 Virgo 上用 3B 重做（791/791 存在）。85 个 PHA 中有 79 个的 GTI 并集与计数事件不同，但 PHA EXPOSURE/事件 LIVETIME 中位数 1.0013，范围 0.987–1.014（在 \(\pm\)11% 响应认证内作为 \(\leq\) 1.4% 接受）；25 个响应门控失败的 PHA/live \(\approx\) 1，因此其 1.26–1.9\(\times\) 超出是面积而非时间。X1b 图轴代理 32 个中 25 个在 1% 内，MOS2 \(\approx\) 0.990，0505760201 mos1 0.858。同一掩护谱系也出现在 S1b → 为 S1b 排入 CORRECTION_01 |
| R131 R131 | 2026-11-05 03:14 EST2026-11-05 03:14 EST | RT-38RT-38 | D1b SFR-shaped soft “gas” termD1b 的 SFR 形状软“气体”项 | WEAKENED → not detected减弱 → 未检出 | Reproduced D1b exactly (\(\Delta\)a \(-\)1e−10). Adding the masked point-source count within 1′ as a covariate cuts the gas amplitude 1.3603 → 0.3673 (\(-\)73 %); cluster bootstrap (7 pointings) SD 0.456 \(\approx\) 21\(\times\) the naive SE, interval [0, 0.97]e−16 includes zero; \(\Sigma_\mathrm{SFR}\) 0–100 Myr fits the residual better than the 0–0.3 Gyr mass (\(\Delta\)nll 848). PSF-wing spill alone does not explain it (1.36 → 1.27). Gas term not detected at 90 %: < 0.97e−16 \(\approx\) L < 1.1e37 erg/s over the footprint (scaled). Its 'local minimum' remark is a nested fit with one more term, not a convergence fault精确复现 D1b（\(\Delta\)a \(-\)1e−10）。把 1′ 内的被掩护点源计数作为协变量加入后，气体幅度从 1.3603 降到 0.3673（\(-\)73%）；簇自助法（7 次取样）SD 0.456 \(\approx\) 朴素 SE 的 21\(\times\)，区间 [0, 0.97]e−16 包含零；\(\Sigma_\mathrm{SFR}\) 0–100 Myr 比 0–0.3 Gyr 质量更好地拟合残余（\(\Delta\)nll 848）。仅 PSF 翼溢出无法解释（1.36 → 1.27）。气体项在 90% 下未检出：< 0.97e−16 \(\approx\) 在该足迹上 L < 1.1e37 erg/s（已标度）。它的“局部极小”说法是多项嵌套拟合，不是收敛故障 |
Source: cognition-os/07_REVISION_LOG.md (R001–R131); cognition-os/04_DECISION_LOG.md (D26–D30).来源：cognition-os/07_REVISION_LOG.md (R001–R131)；cognition-os/04_DECISION_LOG.md (D26–D30)。
Running jobs and what each will decide运行中的任务及其将决定的事项
| Job任务 | Arm / host机臂 / 主机 | Status状态 | Decides when done完成时决定 |
|---|---|---|---|
| D1d age-resolved emissivity: pixel-level support, absolute corner→FOV QPB, D30 input auditD1d 按年龄分辨发射率：像素级支撑、绝对的角区→视场 QPB、D30 输入审计 | sol-dawson high / LTDsol-dawson high / LTD | RUNNING (launched 2026-11-05 02:23 EST; BLOCKED once at the D30 input gate, CORRECTION_01, resumed 2026-11-05 03:14 EST)RUNNING（2026-11-05 02:23 EST 启动；在 D30 输入门控处阻塞一次，CORRECTION_01，2026-11-05 03:14 EST 恢复） | the per-age emissivity with a certified background and response. The chain so far: D1b returned a formal, improved result whose physical gate still failed (R125); D1c fixed the QPB per file from the detector corners but its support rule removed 605/622 regions, leaving 13 regions and 0/24 primary cases passing, so it is formal only and its full-sample row gives old \(\epsilon\) 1.99 [1.72, 2.25]e27 per formed M☉, ratio 82, dev/dof 1.74, QPB pull 8.4 (R129). D1d therefore works at pixel level with an absolute corner\(\rightarrow\)FOV QPB, and its D30 input audit found 61 upstream mask/object images still on rpc/4.background (2.1filter) to be remade from 3B on Virgo, while the PHA GTI differences are \(\leq\) 1.4 % and the 25 response-gate failures are area, not time (R130). When done: accept only if dev/dof < 1.5 and QPB pulls < 3σ.带已认证背景与响应的按年龄发射率。此前的链条：D1b 给出有改进的形式结果，但其物理门控仍失败（R125）；D1c 用探测器角区逐文件固定 QPB，但其支撑规则移除了 605/622 个区域，只剩 13 个区域、24 个主用例中 0 个通过，因此只是形式结果，全样本行给出旧年龄 \(\epsilon\) 1.99 [1.72, 2.25]e27 每已形成 M☉、比值 82、dev/dof 1.74、QPB pull 8.4（R129）。D1d 因此在像素级工作并使用绝对的角区\(\rightarrow\)视场 QPB，其 D30 输入审计发现 61 张上游掩护/目标图像仍在 rpc/4.background（2.1filter）上，需在 Virgo 上用 3B 重做，而 PHA 的 GTI 差异 \(\leq\) 1.4%，25 个响应门控失败是面积而非时间（R130）。完成时：仅在 dev/dof < 1.5 且 QPB pull < 3σ 时接受。 |
| V1b remaining nH profiles (15 of 16)V1b 剩余 nH 剖面（16 个中的 15 个） | codexnim / LTDcodexnim / LTD | RUNNING (resumed 2026-11-04 15:49 EST; \(\leq 2.5\) LTD wall-days)RUNNING（2026-11-04 15:49 EST 恢复；\(\leq 2.5\) 个 LTD 墙钟日） | the profile-likelihood error on the disk plasma term at the halo-field column, which the conditional errors cannot give; the first interval (a024_032) is \(2.15\text{–}2.53\times10^{-7}\), \(2.45 \times\) the conditional error晕场柱密度下圆盘等离子体项的 profile-likelihood 误差，条件性误差给不出它；第一个区间（a024_032）为 \(2.15\text{–}2.53\times10^{-7}\)，是条件性误差的 2.45 倍 |
| RT-37 X1b QPB per sub-bandRT-37 X1b 分波段 QPB | hermeshermes | DONE (2026-11-04 23:39 EST)已完成（2026-11-04 23:39 EST） | answered: the soft QPB scale that broke D1 (R121) is not an energy-dependent response error but the 1.4-2.0 keV Al K\(\alpha\) + Si K\(\alpha\) line band missing from the run6 continuum template — 64.8 % of the corner counts in that band are those lines, and the corner/template ratio is 3.71 there against 0.938 (0.5-0.8 keV) and 1.014 (0.8-1.4 keV) (R127)已有答案：破坏 D1 的软 QPB 尺度（R121）不是能量依赖的响应误差，而是 run6 连续谱模板中缺失的 1.4-2.0 keV Al K\(\alpha\) + Si K\(\alpha\) 线带——该波段角区计数的 64.8% 正是这些线，而角区/模板比值在该波段为 3.71，对比 0.938（0.5-0.8 keV）与 1.014（0.8-1.4 keV）（R127） |
| RT-36 is the nH 0.0585 preference calibration or degeneracy?RT-36 nH 0.0585 的偏好是校准还是退化？ | codexnim / LTD refitscodexnim / LTD 重拟合 | BLOCKED at its 6-h box (2026-11-04 23:39 EST); resumed for the two full refits and per-bin \(\Delta\chi^2\) below 0.5 keV (16-h box)在其 6 小时框处阻塞（2026-11-04 23:39 EST）；为两次完整重拟合与 0.5 keV 以下的逐箱 \(\Delta\chi^2\) 恢复（16 小时框） | answered in part: the 8-annuli preference for \(N_\mathrm{H}=0.0585\) (R116) is driven by E < 0.5 keV — with that band removed the reduced refits prefer 0.03 (\(\Delta\chi^2\) 3.79 and 16.73) — so the preference most likely reflects sub-0.5 keV calibration or soft-sky modelling rather than the column (R126). D1c now uses 0.03 as primary部分已有答案：8 个环带对 \(N_\mathrm{H}=0.0585\) 的偏好（R116）由 E < 0.5 keV 驱动——移除该波段后约化重拟合偏好 0.03（\(\Delta\chi^2\) 3.79 与 16.73）——因此该偏好最可能反映 0.5 keV 以下的校准或软天空建模，而非柱密度本身（R126）。D1c 现在以 0.03 为主 |
| Lit L35 EPIC soft-band calibration and fitted nHLit L35 EPIC 软波段校准与拟合 nH | codexnim / webcodexnim / 网页 | DONE (2026-11-04 23:39 EST)已完成（2026-11-04 23:39 EST） | the nH policy for the foreground-field comparison and the age-resolved regression: published soft residuals are 2-3 % (relative EA) and a 2-3 % energy-dependent residual moves a fitted N_H by ~(0.9-2.5)e20 cm\(^{-2}\); the report "94.5-98.9 % deficit" used 2.85e21 for a 2.85e20 shift, so the V1b shift needs only ~5 % and soft calibration is not excluded (R124)前景视场比较与按年龄分辨回归的 nH 策略：已发表的软波段残余为 2-3 %（相对 EA），2-3 % 的能量依赖残余会把拟合的 N_H 移动 ~(0.9-2.5)e20 cm\(^{-2}\)；报告的“94.5-98.9 % 亏损”对 2.85e20 的移动用了 2.85e21，因此 V1b 的移动只需约 5 %，软波段校准并未被排除（R124） |
| FG1 independent foreground field (Suzaku off-field, XMM mirror, eRASS1)FG1 独立前景视场（Suzaku 离场、XMM 镜像场、eRASS1） | hermes / dispatcher designhermes / 调度方设计 | READY to dispatch (v0 arithmetic done, R120; the full design waits on the nH policy)待派发（v0 算术已完成，R120；完整设计等待 nH 策略） | an independent measurement of the 0.7 keV foreground level: the Suzaku off-field emission measure implies a norm about 0.28–\(0.79 \times\) our halo-field value, but the comparison is provisional until the emission-measure definition and the abundance table are checked0.7 keV 前景水平的独立测量：Suzaku 离场发射率推出的 norm 约为我们晕场值的 0.28–0.79 倍，但在核对发射率定义与丰度表之前该比较只是临时的 |
| S1b: patch the share package to one GTI version (D30)S1b：把共享包修补为单一 GTI 版本（D30） | codexnim / Virgo SAS, lang packagingcodexnim / Virgo SAS，lang 打包 | RUNNING (launched 2026-11-05 01:28 EST; CORRECTION_01 for the mask lineage waits for S1b’s turn end, R130)RUNNING（2026-11-05 01:28 EST 启动；针对掩护谱系的 CORRECTION_01 等待 S1b 轮次结束，R130） | new masked exposure maps, the QPB products, the MANIFEST and a README note on the change, so that every product pairs with the 2.filter events; the X1/F1 forecasts stay superseded by X1b/RT-34. Upload and sharing stay the PI’s. D1d’s input audit found the same 4.background mask lineage inside S1b (R130), so the correction is queued for the package too.新的掩蔽曝光图、QPB 产品、MANIFEST 以及关于该变化的 README 说明，使每个产品都与 2.filter 事件配对；X1/F1 预测仍由 X1b/RT-34 取代。上传与共享仍由 PI 决定。D1d 的输入审计发现 S1b 内部也有同样的 4.background 掩护谱系（R130），因此该更正也对共享包排入。 |
| RT-38 D1b SFR-shaped soft “gas” termRT-38 D1b 的 SFR 形状软“气体”项 | hermeshermes | DONE (2026-11-05 03:14 EST)已完成（2026-11-05 03:14 EST） | answered: the term is not detected. Reproducing D1b exactly, adding the masked point-source count within 1′ as a covariate cuts the gas amplitude 1.3603 → 0.3673 (\(-\)73 %); the cluster bootstrap SD 0.456 is \(\approx\) 21\(\times\) the naive SE and the interval [0, 0.97]e−16 includes zero, so at 90 % the term is < 0.97e−16 \(\approx\) L < 1.1e37 erg/s over the footprint. Its “local minimum” remark is a nested fit with one more term, not a convergence fault (R131).已有答案：该项未检出。精确复现 D1b 后，把 1′ 内的被掩护点源计数作为协变量加入，气体幅度从 1.3603 降到 0.3673（\(-\)73%）；簇自助法的 SD 0.456 \(\approx\) 朴素 SE 的 21\(\times\)，区间 [0, 0.97]e−16 包含零，因此 90% 下该项 < 0.97e−16 \(\approx\) 在该足迹上 L < 1.1e37 erg/s。它的“局部极小”说法是多项嵌套拟合，不是收敛故障（R131）。 |
| RT-39 S1b live-time gate vs the 3B mapsRT-39 S1b 活时间门控 vs 3B 图 | hermes / lang lighthermes / lang 轻量 | RUNNING (launched 2026-11-05 03:14 EST)RUNNING（2026-11-05 03:14 EST 启动） | recompute one failing row’s map EXPOSURE against its event STDGTI/LIVETIME by hand, which sets the S1b rebuild scope.手工重算一行失败案例的图 EXPOSURE 与其事件 STDGTI/LIVETIME，据此确定 S1b 重建的范围。 |
| Lit L37 hot-ISM L_X per SFRLit L37 热 ISM 的每 SFR L_X | codexnim / webcodexnim / 网页 | RUNNING (launched 2026-11-04 23:40 EST)RUNNING（2026-11-04 23:40 EST 启动） | the published hot-ISM luminosity per star-formation rate, checked at its source, which feeds the gas row.已发表的热 ISM 每恒星形成率光度，在其来源处核对，这为气体行提供输入。 |
| Handbook HB01 stage 4d refreshHandbook HB01 第 4d 阶段刷新 | hermes / handbookhermes / 手册 | RUNNING (stage 4d, this refresh)RUNNING（第 4d 阶段，本次刷新） | review-ready → the dispatcher publishes to an existing Pages projectreview-ready → 调度方发布到现有 Pages 项目 |
Source: cognition-os/08_RUNNING.md §3 and §5 (revision 49, 2026-11-05 03:56 EST); 07_REVISION_LOG.md R115, R116, R118, R120, R121, R122, R123, R124, R125, R126, R127, R128, R129, R130, R131; 04_DECISION_LOG.md D30.来源：cognition-os/08_RUNNING.md §3 与 §5（修订版 49，2026-11-05 03:56 EST）；07_REVISION_LOG.md R115、R116、R118、R120、R121、R122、R123、R124、R125、R126、R127、R128、R129、R130、R131；04_DECISION_LOG.md D30。
The one-foreground story单一前景的故事
The question a student would ask: M31 is 2.5 million light-years away, so how could our own Galaxy be responsible for light coming from its disk?
学生可能会问的问题：M31 在 2.5 百万光年之外，那么我们自己的银河系怎么会对来自它的圆盘的光负责？
The physical picture. When you look at M31, you look through the Milky Way first. Along that line of sight sit several layers of cool gas: the Local Hot Bubble (our own bubble of hot gas, about \(\sim\)0.1 keV), the Milky Way's own diffuse halo (about 0.25 keV), and a hotter Galactic halo component near 0.7 keV. Each emits soft X-rays that the detector cannot tell apart from X-rays that really came from M31. On top of that, the Milky Way's own unresolved stars — ordinary stars like the Sun, but with magnetic coronae hot enough to shine in X-rays — add a soft, unstructured glow. So the sky in the direction of M31 is never black; it has a floor.
物理图像。你看 M31 时，是穿过银河系在看它。这条视线上有若干层冷气体：我们自己的热局部泡（Local Hot Bubble，约 \(\sim\)0.1 keV）、银河系自身的弥散晕（约 0.25 keV），以及一个约 0.7 keV 的更热的银河系晕成分。每一层都发出软 X 射线，而探测器无法把它们与真正来自 M31 的 X 射线区分开。除此之外，银河系自身未解析的恒星——像太阳一样的普通恒星，但磁日冕热到能在 X 射线波段发光——又加上一层柔和、无结构的辉光。所以指向 M31 的天空从来不是黑的；它有一个底板。
Why this matters here. The disk fits switch the Milky Way's 0.7 keV halo term off (they set it to zero) and then fit a plasma component, SrcApec, to whatever soft emission is left over. If the Milky Way's own 0.7 keV glow is really there, it is not being fitted as a foreground — it is being absorbed into the M31 component. The observed-to-predicted ratio then rises with radius not because M31's gas grows, but because the foreground was left out of the model in the first place.
为什么这对本项目重要。圆盘拟合把银河系 0.7 keV 晕项关掉（设为零），然后用一个等离子体成分 SrcApec 去拟合剩下的软辐射。如果银河系自身的 0.7 keV 辉光确实存在，它并没有被当作前景拟合——而是被吸进了 M31 的成分里。于是观测/预测比随半径上升，不是因为 M31 的气体变多了，而是因为前景从一开始就没进模型。
S_unabs_errB, the stored statistical error on the
unabsorbed 0.5–2 keV surface brightness.S_unabs_errB，
即未吸收 0.5–2 keV 表面亮度的存储统计误差。Source: data/executor-results/2026-11-02-v10-e045-opus/tables/e045_hard_soft_per_annulus.csv (disk SB, S_unabs, S_unabs_errB, sigma_star, S_pred_M32); data/executor-results/2026-11-02-rt10-sonnet/tables/obj1_hothalo_per_field.csv (halo-field SB052); 07_REVISION_LOG.md R024, R027, R032; draft/insights/I021_one_07kev_foreground_three_numbers.md. Script: handbook/scripts/make_figures.py (figure_i021).来源：data/executor-results/2026-11-02-v10-e045-opus/tables/e045_hard_soft_per_annulus.csv（圆盘 SB、S_unabs、S_unabs_errB、sigma_star、S_pred_M32）；data/executor-results/2026-11-02-rt10-sonnet/tables/obj1_hothalo_per_field.csv（晕场 SB052）；07_REVISION_LOG.md R024、R027、R032；draft/insights/I021_one_07kev_foreground_three_numbers.md。脚本：handbook/scripts/make_figures.py（figure_i021）。
Source: .agent/jobs/2026-11-01-v2-e036-e033-branchc-kt-profile-grid/profile.csv (the 13-point grid and the null statistic); data/executor-results/2026-11-04-rt31-halo-profile-convergence-codexnim/rt31_summary.csv and .../resume01/resume01_summary.json (the warm restarts, q and the 90 % norm interval); 07_REVISION_LOG.md R071, R079, R110, R112, R093, R027, R086; .agent/jobs/2026-11-03-redteam-rt26-halo-level-after-v2b-norm/results/rt26_results2.json (the two estimator levels and their separation). Script: handbook/scripts/make_figures_halo.py.来源：.agent/jobs/2026-11-01-v2-e036-e033-branchc-kt-profile-grid/profile.csv（13 点网格与零假设统计量）；data/executor-results/2026-11-04-rt31-halo-profile-convergence-codexnim/rt31_summary.csv 与 .../resume01/resume01_summary.json（热重启、q 与 90% norm 区间）；07_REVISION_LOG.md R071、R079、R110、R112、R093、R027、R086；.agent/jobs/2026-11-03-redteam-rt26-halo-level-after-v2b-norm/results/rt26_results2.json（两个估计器水平及其间距）。脚本：handbook/scripts/make_figures_halo.py。
The measurement that answers it. Three numbers should agree if the story is right. The halo fields (20 XMM fields 20–100′ away from M31, where no M31 gas should be) give one level of 0.7 keV emission: \(0.97 \pm 0.12\times10^{-7}\) arcmin\(^{-2}\), \(8\sigma\) above zero, and flat in distance from M31. The disk's soft floor, converted to the same units, is \(1.13 \pm 0.08\times10^{-7}\) — \(1.1\sigma\) away. And the Milky Way stellar prediction from the literature is \(0.72\)–\(1.05\times10^{-7}\) bright-removed. All three land on the same number. The published HaloSat Milky Way hot component toward M31 is \(3.28\times10^{-16}\) against our \(2.9\times10^{-16}\), so a Galactic component of this size is expected.
回答它的测量。如果这个故事成立，三个数字应当一致。晕场（20 个距 M31 20–100′ 的 XMM 视场，那里不应有 M31 气体）给出一个统一的 0.7 keV 辐射水平：\(0.97 \pm 0.12\times10^{-7}\) arcmin\(^{-2}\)， 距零 \(8\sigma\)，且与到 M31 的距离无关。圆盘的软底板换成同一体例是 \(1.13 \pm 0.08\times10^{-7}\)——相差 \(1.1\sigma\)。 而文献给出的银河系恒星预测是 \(0.72\)–\(1.05\times10^{-7}\)（亮源移除情形）。三个数字落在同一个数上。 已发表的 HaloSat 指向 M31 的银河系热成分是 \(3.28\times10^{-16}\)，对比我们的 \(2.9\times10^{-16}\)， 所以这个量级的银河系成分是意料之中的。
What is still open. The story is Provisional, and since 2026-11-03 it has two separate open ends. First, the halo fields do contain a 0.7 keV component: the registered full-freedom null rejects it at \(q = 22.4\) and six warm restarts push that to \(q = 34.8\), with the component at \(kT \le 0.80\) keV (90 %) and a norm of \(3.2\text{–}5.8\times10^{-8}\) arcmin\(^{-2}\). What that measurement does not say is whether the component is the Milky Way or M31's circumgalactic medium — the registered set-up switches the Milky Way 0.7 keV term off, so the fitted component is the halo fields' foreground by construction. Second, the level itself turned out to be estimator-dependent: the same data give \(9.66 \pm 1.16\times10^{-8}\) in one fit and \(5.0 [3.56, 7.43]\times10^{-8}\) in another, \(2.3\text{–}2.7\sigma\) apart, so "one common level behind the halo fields, the disk floor and the stellar prediction" is a model-dependent coincidence (Questioned). The outer-disk rule came back undecided (a SrcApec of zero with an adequate fit is not evidence that the foreground exceeds the total), and a red team weakened the working conclusion to "current fits do not require diffuse M31 gas; the emission is consistent with stellar-like emission plus an extended foreground mixture". Inside the foreground, stars versus gas is still open.
The second number that changed. On 2026-11-03 the halo-field level was quoted as one number, \(0.97 \pm 0.12\times10^{-7}\) arcmin\(^{-2}\). Three days of fits later the same quantity is quoted as a range, because two registered fits of the same data disagree by \(2.3\text{–}2.7\sigma\). Nothing about the sky changed; the estimators did. The lesson is the same one the 4.2-sigma correction taught: a level is only as well-defined as the estimator that produced it, so the page now quotes the range and names both estimators.
第二个变了的数字。2026-11-03，晕场水平被引用为一个数字，\(0.97 \pm 0.12\times10^{-7}\) arcmin\(^{-2}\)。三天拟合之后，同一量被引用为一个区间，因为同一份数据的两个已注册拟合相差 \(2.3\text{–}2.7\sigma\)。天空没有变化；变化的是估计器。教训与 4.2 sigma 那次修正相同：一个水平只有在其产生它的 估计器一样明确时才明确，所以本页现在引用区间并点名两个估计器。
A number that changed, and why it matters. In the middle of the day the project published a prediction that transplanting the halo-field level onto the outer disk would over-subtract by \(4.2\sigma\), which would have argued against the story. A red-team replay showed the prediction had used the wrong error bar: it quoted a conditional error where the decision rule names a scaled one. With the registered error the transplant gives an intercept of \(-1.25 \pm 1.04\times10^{-16}\), only \(1.2\sigma\) from zero — so the arithmetic no longer decides, and the refits do. This is the normal kind of correction in this project: a number is published, attacked, and revised with the error bar named in advance. When you read a number here, always read which error was used.
仍未解决之处。这个故事是临时性的，而自 2026-11-03 起它有两个分开的 开放端。第一，晕场确实含有一个 0.7 keV 成分：已注册的全自由度零假设以 \(q = 22.4\) 拒绝它，六次热重启把 这一数字推到 \(q = 34.8\)，该成分在 \(kT \le 0.80\) keV（90%），norm 为 \(3.2\text{–}5.8\times10^{-8}\) arcmin\(^{-2}\)。但这项测量没有说明该成分是银河系还是 M31 的星周介质——已注册的设置关闭了银河系 0.7 keV 项，因此被拟合的成分按构造就是晕场的前景。第二，水平本身被证明依赖于估计器：同一份数据在一个拟合中给出 \(9.66 \pm 1.16\times10^{-8}\)，在另一个中给出 \(5.0 [3.56, 7.43]\times10^{-8}\)，相差 \(2.3\text{–}2.7\sigma\)， 所以“晕场、圆盘底板与恒星预测背后有一个共同水平”是模型依赖的巧合（受质疑）。外圆盘规则给出的裁定是 未决（SrcApec 为零且拟合合格，并不是前景超过总量的证据），一次红队把工作结论减弱为“当前拟合不需要 弥散的 M31 气体；发射与恒星状发射加上一个扩展的前景混合一致”。在前景内部，恒星还是气体仍然开放。
一个变了的数字，以及它为什么重要。当天中途，项目发布过一个预测：把晕场水平移植到外圆盘会过度扣除 \(4.2\sigma\)，那本会不利于这个故事。一次红队重放发现，该预测用错了误差棒——它引用了条件性误差，而决策规则指定的是缩放后的误差。用注册误差，移植给出截距 \(-1.25 \pm 1.04\times10^{-16}\)，距零仅 \(1.2\sigma\)——于是算术不再决定问题，由重拟合决定。 这就是本项目中正常的修正方式：一个数字被发布、被攻击，再用事先指定的误差棒修订。在这里读数字时，务必同时读它用的是哪一种误差。
Source: cognition-os/draft/insights/I021_one_07kev_foreground_three_numbers.md (label Provisional, R032); cognition-os/07_REVISION_LOG.md R024, R027, R029, R030, R032, R033, R034, R050, R051, R053, R071, R079, R080, R093, R110, R112, R115; cognition-os/03_TRUST_TABLE.md (row "One 0.7 keV foreground behind the halo-field term, the H1 MW-star prediction and the disk soft floor" — one common level Questioned, Medium).来源：cognition-os/draft/insights/I021_one_07kev_foreground_three_numbers.md（标签 Provisional，R032）；cognition-os/07_REVISION_LOG.md R024、R027、R029、R030、R032、R033、R034、R050、R051、R053、R071、R079、R080、R093、R110、R112、R115；cognition-os/03_TRUST_TABLE.md（行“一个 0.7 keV 前景，位于晕场项、H1 MW 恒星预测和圆盘软底板之后”——共同水平受质疑，中等）。
How sure are we of each我们对每一事项有多确信
Trust labels are the project's own, recorded in 03_TRUST_TABLE.md and changed through 07_REVISION_LOG.md. A label is a statement about the strength of the evidence behind a claim, not a verdict of right/wrong. Rows marked Questioned, Withdrawn or Retracted must not be cited as findings. R016 (revision 14) contradicts the frozen disk nH=0.03 magnitude: the MW-only foreground column is \(\approx\)0.054–0.067e22, about twice the frozen value; V1's bracket 0.048/0.0585/0.076 stands. R042 then measured the effect in one annulus: at the halo-field column 0.0585e22 the disk plasma component rises 4.8 % in intrinsic surface brightness and the Milky Way halo term rises 18 % (9.6\(\sigma\)), so the other eight annuli are being refitted. R043, R048 and R049 then narrowed the inner-disk row three times: the excess survives the measured point-spread spill and the particle-background systematics, but M31's own gas along the line of sight can soften the soft band only in the outer annuli. A label can go Provisional \(\rightarrow\) Questioned \(\rightarrow\) Provisional again inside one day (R047, then R048); always read the date on the row. Since then the labels have moved twice more: the diffuse-gas row is now Provisional with an upper limit rather than a detection (R064), and the one-foreground hypothesis is still Questioned as to its origin because the decomposition systematics are as large as the effect (R057). One premise was withdrawn outright: the accepted annulus fits do not rest on the run6 campaign (R070). The labels have moved three more times since: the halo-field component is now detected (R071, R079, R110, R112) but its attribution is not (the Milky Way 0.7 keV term is switched off in the registered set-up, so the fitted component is the halo fields' foreground by construction); the "one common level" behind the halo fields and the disk floor is Questioned because the level is estimator-dependent (R093); and the disk gas limit became conditional rather than a single number when the floor was fitted instead of borrowed (R091, R098).
信任标签是项目自身的，记录在 03_TRUST_TABLE.md 中，并通过 07_REVISION_LOG.md 更改。标签是对支撑一项声明的证据强度的陈述，而非对错的裁决。标记为 Questioned、Withdrawn 或 Retracted 的行不得作为发现引用。R016（修订版 14）否定了冻结圆盘 nH=0.03 的量级：仅 MW 前景柱密度 \(\approx\)0.054–0.067e22，约为冻结值的两倍；V1 的括号 0.048/0.0585/0.076 保持。 随后 R042 在一个环带上量出了它的影响：在晕场柱密度 0.0585e22 下，圆盘等离子体成分的本征表面亮度上升 4.8 %，银河系晕项上升 18 %（9.6\(\sigma\)），其余八个环带正在重拟合。接着 R043、R048、R049 三次收窄了内圆盘那一行：超额在量出的点扩散溢出和粒子背景系统误差之后仍然存在，但 M31 自身视线方向上的气体只能软化外环带的软波段。一个标签可以在一天之内 Provisional \(\rightarrow\) Questioned \(\rightarrow\) 再次 Provisional（R047，然后 R048）；所以务必读该行的日期。此后标签又移动了两次：弥散气体那一行现在是带上限的临时性，而非检出（R064）；单一前景假设就其起源而言仍受质疑，因为分解系统误差与效应本身一样大（R057）。还有一条前提被完全撤回：受理的环带拟合并不建立在 run6 运动之上（R070）。此后标签又移动了三次：晕场成分现在是已检出（R071、R079、R110、R112），但其归因尚未确定（已注册的设置中银河系 0.7 keV 项被关闭，因此被拟合的成分按构造就是晕场的前景）；“晕场与圆盘底板背后有一个共同水平”受质疑，因为该水平依赖于估计器（R093）；当底板被拟合而非借用时，圆盘气体上限从单一数字变为有条件（R091、R098）。
数据截至：08 revision 49Data as of: 08 revision 49
Newest source read: cognition-os/08_RUNNING.md, revision 49, updated 2026-11-05 03:56 EST; cognition-os/03_TRUST_TABLE.md read 2026-11-05 03:15 EST; 07_REVISION_LOG.md rows R001–R131 and 04_DECISION_LOG.md D26–D30 read in full.
最新读取的来源：cognition-os/08_RUNNING.md，修订版 49，更新于 2026-11-05 03:56 EST；cognition-os/03_TRUST_TABLE.md 读取于 2026-11-05 03:15 EST；07_REVISION_LOG.md 的 R001–R131 行与 04_DECISION_LOG.md 的 D26–D30 已全文读取。