Publish codex SIMPLE seed-0 evaluation evidence

data.json

@@ -6,7 +6,7 @@
   "benchmark_complete": false,
   "edition": "codex-simple-astra-high-seed0",
   "created_at": "2026-10-09T00:09:42.989838+00:00",
-  "updated_at": "2026-10-10T01:36:44.778308+00:00",
+  "updated_at": "2026-10-10T01:57:37.536849+00:00",
   "model": "gpt-6-astra",
   "effort": "high",
   "seed": 0,
@@ -138,23 +138,17 @@
     ],
     "interrupted_attempts": 4,
     "usage_incomplete_tasks": [],
-    "execution_incomplete_tasks": [
-      "task06/09",
-      "task06/10",
-      "task06/11"
-    ],
+    "execution_incomplete_tasks": [],
     "deferred_tasks": [],
     "new_evaluations": 8,
     "progress": {
       "finished": 80,
-      "running": 0,
-      "queued": 0,
+      "running": 3,
+      "queued": 1,
       "needs_review": 0,
-      "interrupted": 3,
+      "interrupted": 0,
       "native_successes": 51,
-      "native_failures": 29,
-      "not_started": 1,
-      "stopped_by_user": true
+      "native_failures": 29
     },
     "simple_campaign": {
       "planned": 12,
@@ -6691,8 +6685,8 @@
       "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
       "status": "pending",
       "episode_id": null,
-      "run_status": "interrupted",
-      "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+      "run_status": "running",
+      "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
       "planned_protocol": {
         "episodes": 1,
         "seed": 0,
@@ -6735,8 +6729,8 @@
       "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
       "status": "pending",
       "episode_id": null,
-      "run_status": "interrupted",
-      "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+      "run_status": "running",
+      "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
       "planned_protocol": {
         "episodes": 1,
         "seed": 0,
@@ -6779,8 +6773,8 @@
       "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
       "status": "pending",
       "episode_id": null,
-      "run_status": "interrupted",
-      "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+      "run_status": "running",
+      "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
       "planned_protocol": {
         "episodes": 1,
         "seed": 0,
@@ -6823,8 +6817,8 @@
       "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
       "status": "pending",
       "episode_id": null,
-      "run_status": "not_started",
-      "status_note": "Stopped at the user request before this task started.",
+      "run_status": "queued",
+      "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
       "planned_protocol": {
         "episodes": 1,
         "seed": 0,
@@ -27487,14 +27481,5 @@
   "instruction_review": {
     "policy": "Lift pot uses native RoboTwin TCP frames after the RoboEnv correction, with the existing reviewed instruction, unchanged native verifier, seed 0, 7500 action frames at 25 Hz, and one fresh stock Codex 0.160.0 / GPT-6 Astra high episode. Other results retain their historical runtime and protocol."
   },
-  "authorized_selection_complete": false,
-  "execution_stop": {
-    "time": "2026-10-10T01:32:09.008817+00:00",
-    "reason": "User confirmed the runs were stopped from another session and requested keeping them stopped; finish Space publication only.",
-    "dispatch": false,
-    "resume_or_replace_episodes": false,
-    "completed_results_retained": 16,
-    "interrupted_attempts_retained": 7,
-    "never_started": 1
-  }
+  "authorized_selection_complete": false
 }

manifest.json

@@ -19,7 +19,7 @@
   "catalog/task06/11-push-office-chair/task.yaml": "0e2de1c50619b0c58fd939f58362dfc0630f34d4e19587fe4927a2c4e5c1d785",
   "catalog/task06/12-open-trash-can/task.yaml": "e02833927892caa9c40ed193348aab1858e6ae6208c4c04892dec8eae4ad45f0",
   "comparison.json": "55202006828502b083284dee5a70ad74a8af369afa34e258419ddb5fd2702c01",
-  "data.json": "6be970685887538176ef38cfdd0b01b3108316712bb02b588d4daae9070b054e",
+  "data.json": "33b5672e195b45063a7322f7dd8695272aef6debf25349b66e08876fe18a7376",
   "episodes.csv": "7f4d5f46d65689ff303609a98ffa8a34ad875e7cc9b6018628fc368e87070ec9",
   "episodes.json": "41c1c467c7c36745454b1f1da54e2cbd25823af225581cfa168472878f4ea0f9",
   "failure-reviews.json": "c66a997c9ecd6f7f43f130a8a33cd64c7a5b2049033e9ccde653ee30e4796f23",
@@ -32,6 +32,6 @@
   "licenses/robosuite-LICENSE.txt": "177978cbece0a4c454c2aaec5b3f145b39270814874c43109da9e829c39d9cba",
   "protocol.json": "205b330174d7930ec9d503ffd9f8d957716a5e72357902ba3352b09bd3824781",
   "style.css": "dd31b30951c220ebb82bc96224b5e640913945679e42b2e0a42e2c373a329e6c",
-  "summary.json": "d0c28a28c57588bc7cf5f244a20a82cd15d621fb3d390534daa72e24a45c1ec8",
-  "task-index.json": "e915eae26307309d95c71da154a5b8cad575b26768e4da97220a28c498a0c3b4"
+  "summary.json": "e98375ff78d7f147e909a20fd3f265865e416c3d57ae045ded61a3d0ba17bf1b",
+  "task-index.json": "4c10062dfd461b34d04ab2e967c0cc6f683f61872409b3c4f44ba20eb6cd2379"
 }

publication-manifest.json

@@ -82,8 +82,8 @@
       "bytes": 134846
     },
     "data.json": {
-      "sha256": "6be970685887538176ef38cfdd0b01b3108316712bb02b588d4daae9070b054e",
-      "bytes": 1172466
+      "sha256": "33b5672e195b45063a7322f7dd8695272aef6debf25349b66e08876fe18a7376",
+      "bytes": 1171722
     },
     "episodes.csv": {
       "sha256": "7f4d5f46d65689ff303609a98ffa8a34ad875e7cc9b6018628fc368e87070ec9",
@@ -134,15 +134,15 @@
       "bytes": 16560
     },
     "summary.json": {
-      "sha256": "d0c28a28c57588bc7cf5f244a20a82cd15d621fb3d390534daa72e24a45c1ec8",
-      "bytes": 3693
+      "sha256": "e98375ff78d7f147e909a20fd3f265865e416c3d57ae045ded61a3d0ba17bf1b",
+      "bytes": 3589
     },
     "task-index.json": {
-      "sha256": "e915eae26307309d95c71da154a5b8cad575b26768e4da97220a28c498a0c3b4",
-      "bytes": 234598
+      "sha256": "4c10062dfd461b34d04ab2e967c0cc6f683f61872409b3c4f44ba20eb6cd2379",
+      "bytes": 234352
     },
     "manifest.json": {
-      "sha256": "bd84d8f0740ae7bc81fae5287e36017c3acede88fa10da71bcdd3ee698dbd26d",
+      "sha256": "7f35d9143b9b5c3b2894f27df5f9b0971a92b4dbd1300c2b19a403d62a5f6341",
       "bytes": 3481
     }
   }

summary.json

@@ -121,23 +121,17 @@
   ],
   "interrupted_attempts": 4,
   "usage_incomplete_tasks": [],
-  "execution_incomplete_tasks": [
-    "task06/09",
-    "task06/10",
-    "task06/11"
-  ],
+  "execution_incomplete_tasks": [],
   "deferred_tasks": [],
   "new_evaluations": 8,
   "progress": {
     "finished": 80,
-    "running": 0,
-    "queued": 0,
+    "running": 3,
+    "queued": 1,
     "needs_review": 0,
-    "interrupted": 3,
+    "interrupted": 0,
     "native_successes": 51,
-    "native_failures": 29,
-    "not_started": 1,
-    "stopped_by_user": true
+    "native_failures": 29
   },
   "simple_campaign": {
     "planned": 12,

task-index.json

@@ -6406,8 +6406,8 @@
     "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
     "status": "pending",
     "episode_id": null,
-    "run_status": "interrupted",
-    "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+    "run_status": "running",
+    "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
     "planned_protocol": {
       "episodes": 1,
       "seed": 0,
@@ -6450,8 +6450,8 @@
     "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
     "status": "pending",
     "episode_id": null,
-    "run_status": "interrupted",
-    "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+    "run_status": "running",
+    "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
     "planned_protocol": {
       "episodes": 1,
       "seed": 0,
@@ -6494,8 +6494,8 @@
     "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
     "status": "pending",
     "episode_id": null,
-    "run_status": "interrupted",
-    "status_note": "The prior attempt was interrupted at the user request; no valid native verdict. Execution remains stopped. Retained separately from native failures.",
+    "run_status": "running",
+    "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
     "planned_protocol": {
       "episodes": 1,
       "seed": 0,
@@ -6538,8 +6538,8 @@
     "instruction_source": "Unchanged native SIMPLE instruction for the effective default task configuration.",
     "status": "pending",
     "episode_id": null,
-    "run_status": "not_started",
-    "status_note": "Stopped at the user request before this task started.",
+    "run_status": "queued",
+    "status_note": "Authorized episode is queued, running, or awaiting evidence review.",
     "planned_protocol": {
       "episodes": 1,
       "seed": 0,