{"abstract":"Gradient bucket reconstruction discards global offsets.","category":"Distributed training aggregation","checks":6,"contract":"Given [offset,values] and total length, reconstruct exactly covered list or return None for gaps/overlaps.","contract_signature":"chunks, total","evaluation_group":"w3-dta-bucket-boundary","failed_approach":"Appending payloads by width can put values into the wrong bucket.","family":"w3-distributed-training-aggregation-bucket-boundary","id":"FA-99326","implementations":{"attempt":{"sha256":"70c688f564fa364063ee6aac95dad4e433548a25db92031822a25aab49d2bd8f","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(chunks, total):\n    out=[None]*total\n    for s,vs in chunks:\n        for j,v in enumerate(vs):\n            if 0<=s+j<total: out[s+j]=v\n    return out\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('chunks reversed',solve([[2,[2,3]],[0,[0,1]]],4),[0,1,2,3])\ncheck('overlap rejected',solve([[0,[0,1,2]],[2,[2,3]]],4),None)\ncheck('gap rejected',solve([[0,[0]],[2,[2,3]]],4),None)\ncheck('empty tensor',solve([],0),[])\ncheck('singleton chunk',solve([[0,[N]]],1),[N])\ncheck('out of bounds',solve([[0,[0,1]]],1),None)\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"},"broken":{"sha256":"c0403751c0b27211a4355a4c30e1be47cd9b435e385ee369b15bb4e7a28c5a78","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(chunks, total):\n    return [v for s,vs in chunks for v in vs]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('chunks reversed',solve([[2,[2,3]],[0,[0,1]]],4),[0,1,2,3])\ncheck('overlap rejected',solve([[0,[0,1,2]],[2,[2,3]]],4),None)\ncheck('gap rejected',solve([[0,[0]],[2,[2,3]]],4),None)\ncheck('empty tensor',solve([],0),[])\ncheck('singleton chunk',solve([[0,[N]]],1),[N])\ncheck('out of bounds',solve([[0,[0,1]]],1),None)\nprint(json.dumps({\"observations\": observations, \"passed\": all(x[\"passed\"] for x in observations)}, ensure_ascii=False))\nraise SystemExit(0 if all(x[\"passed\"] for x in observations) else 1)\n"}},"limitations":"Inputs are already materialized Python values from a fixed toy collective; no transport, failure recovery, optimizer state, or real model behavior is represented. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.","method":"Deterministic executable model with adversarial boundary fixtures.","provenance":{"created_by":"Failure Map","dependencies":"Python standard library","family":"w3-distributed-training-aggregation-bucket-boundary","generated_at":"2026-09-30T16:48:23.079335+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A toy distributed-training contract makes replica and shard contribution rules explicit without claiming framework compatibility or performance.","root_cause":"Different workers may contribute different bucket widths; bucket ownership is defined by global offsets.","sha256":"1a8f623a0869cdda428fd14b17c2f8b2f0c97489046d5da23a89631fc1418929","title":"Gradient bucket reconstruction discards global offsets · case 01","variant":1,"variant_policy":"The fixture code references N, supplied as 1 through 5. Fixed boundary checks may repeat. These records share one mechanism and must remain together in evaluation splits.","verified":true,"visibility":"public","verification":{"attempt":{"elapsed_ms":27.354,"exit_code":1,"observations":[{"actual":[0,1,2,3],"check":"chunks reversed","expected":[0,1,2,3],"passed":true},{"actual":[0,1,2,3],"check":"overlap rejected","expected":null,"passed":false},{"actual":[0,null,2,3],"check":"gap rejected","expected":null,"passed":false},{"actual":[],"check":"empty tensor","expected":[],"passed":true},{"actual":[1],"check":"singleton chunk","expected":[1],"passed":true},{"actual":[0],"check":"out of bounds","expected":null,"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"chunks reversed\", \"actual\": [0, 1, 2, 3], \"expected\": [0, 1, 2, 3], \"passed\": true}, {\"check\": \"overlap rejected\", \"actual\": [0, 1, 2, 3], \"expected\": null, \"passed\": false}, {\"check\": \"gap rejected\", \"actual\": [0, null, 2, 3], \"expected\": null, \"passed\": false}, {\"check\": \"empty tensor\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton chunk\", \"actual\": [1], \"expected\": [1], \"passed\": true}, {\"check\": \"out of bounds\", \"actual\": [0], \"expected\": null, \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":26.682,"exit_code":1,"observations":[{"actual":[2,3,0,1],"check":"chunks reversed","expected":[0,1,2,3],"passed":false},{"actual":[0,1,2,2,3],"check":"overlap rejected","expected":null,"passed":false},{"actual":[0,2,3],"check":"gap rejected","expected":null,"passed":false},{"actual":[],"check":"empty tensor","expected":[],"passed":true},{"actual":[1],"check":"singleton chunk","expected":[1],"passed":true},{"actual":[0,1],"check":"out of bounds","expected":null,"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"chunks reversed\", \"actual\": [2, 3, 0, 1], \"expected\": [0, 1, 2, 3], \"passed\": false}, {\"check\": \"overlap rejected\", \"actual\": [0, 1, 2, 2, 3], \"expected\": null, \"passed\": false}, {\"check\": \"gap rejected\", \"actual\": [0, 2, 3], \"expected\": null, \"passed\": false}, {\"check\": \"empty tensor\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton chunk\", \"actual\": [1], \"expected\": [1], \"passed\": true}, {\"check\": \"out of bounds\", \"actual\": [0, 1], \"expected\": null, \"passed\": false}], \"passed\": false}\n"}},"member_only":{"stages":["fixed"],"fields":["implementations.fixed","verification.fixed","harness","repair"],"note":"The verified repair, its recorded checks, the repair description, and the scoring harness are available to members."}}