Auto-sync: 2026-06-28 13:38:23

results/README.md

@@ -4,6 +4,8 @@ Use these summaries for the current paper tables:
 
 - `paper_story_memo.md`
 - `paper_table_status.md` and `.json` (generated by `scripts/build_paper_table_status.py`)
+- `paper_analysis.md` and `.json` (generated by `scripts/build_paper_analysis.py`;
+  seed-wise paired deltas, per-task gaps, and selection histograms)
 - `nonexpert_proposal_target_census.md`
 - `field_selected_target_map_summary.md`
 - `h16_field_sweep_summary.md`
@@ -89,7 +91,7 @@ Completed deployment-clean diagnostics:
 
 Completed residual transport jobs:
 
-Last checked: `2026-06-28 17:28 UTC`. Ray-search jobs are queued, pending on
+Last checked: `2026-06-28 17:36 UTC`. Ray-search jobs are queued, pending on
 GPU priority.
 
 - Code supports `--retrieval-residual-scale`, defaulting to `1.0` so the
@@ -141,7 +143,8 @@ GPU priority.
   `0.40` both reach 34.67%. K2 raw residual transport remains the best clean row.
 - Jobs `14868993`/`14868995`/`14868997`/`14868999` are pending counterfactual
   tangent ray-search rollouts. Summary jobs `14868994`/`14868996`/`14868998`/
-  `14869000` and table rebuild job `14869605` are dependency-gated on them.
+  `14869000`, table rebuild job `14869605`, and updated table+analysis rebuild
+  job `14869860` are dependency-gated on them.
 - Job `14869627` completed the CPU Apptainer smoke for the residual scale-grid
   selector: selected index `3` and returned action `0.20` on the toy ray-search
   case, validating candidate expansion before rollout.
@@ -160,4 +163,4 @@ Paper table automation:
   existing summary JSON files, uses canonical fallback values for already
   verified rows, and marks pending experiments by job ID.
 - `scripts/slurm/build_paper_table_status.sbatch` reruns that table builder on a
-  CPU node with system `python3`.
+  CPU node with system `python3`, then regenerates `paper_analysis.*`.