{"abstract":"All-reduce silently truncates vectors with mismatched shapes.","category":"Distributed training aggregation","checks":6,"contract":"Given worker vectors, return elementwise sum if all lengths match, otherwise None; empty input returns [].","contract_signature":"parts","evaluation_group":"w3-dta-allreduce-sum-shape","failed_approach":"Padding short vectors with zeros still accepts an invalid collective shape.","family":"w3-distributed-training-aggregation-allreduce-sum-shape","id":"FA-99306","implementations":{"attempt":{"sha256":"2371146cbbdcf6e24156500196090cf2ffd1035720b0515b460e616c42f7d58f","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(parts):\n    w=max(map(len,parts),default=0); return [sum(v[j] if j<len(v) else 0 for v in parts) for j in range(w)]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('equal shape',solve([[N,2],[3*N,4]]),[4*N,6])\ncheck('mismatch rejected',solve([[N,2],[3*N]]),None)\ncheck('empty',solve([]),[])\ncheck('singleton empty vector',solve([[]]),[])\ncheck('singleton values',solve([[N]]),[N])\ncheck('all empty shapes',solve([[],[]]),[])\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":"b34df758a386c986fec44e8a555796828acfade0bb876cdc11462e2ce976ad1d","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(parts):\n    return [sum(col) for col in zip(*parts)] if parts else []\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('equal shape',solve([[N,2],[3*N,4]]),[4*N,6])\ncheck('mismatch rejected',solve([[N,2],[3*N]]),None)\ncheck('empty',solve([]),[])\ncheck('singleton empty vector',solve([[]]),[])\ncheck('singleton values',solve([[N]]),[N])\ncheck('all empty shapes',solve([[],[]]),[])\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-allreduce-sum-shape","generated_at":"2026-09-30T16:48:22.633481+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":"Zip-based summation stops at the shortest worker vector and hides malformed contributions.","sha256":"7ff9736fec231603ffcf898154cccf5653f06d543440608aa65f79ea3fadde11","title":"All-reduce silently truncates vectors with mismatched shapes · 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":28.654,"exit_code":1,"observations":[{"actual":[4,6],"check":"equal shape","expected":[4,6],"passed":true},{"actual":[4,2],"check":"mismatch rejected","expected":null,"passed":false},{"actual":[],"check":"empty","expected":[],"passed":true},{"actual":[],"check":"singleton empty vector","expected":[],"passed":true},{"actual":[1],"check":"singleton values","expected":[1],"passed":true},{"actual":[],"check":"all empty shapes","expected":[],"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"equal shape\", \"actual\": [4, 6], \"expected\": [4, 6], \"passed\": true}, {\"check\": \"mismatch rejected\", \"actual\": [4, 2], \"expected\": null, \"passed\": false}, {\"check\": \"empty\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton empty vector\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton values\", \"actual\": [1], \"expected\": [1], \"passed\": true}, {\"check\": \"all empty shapes\", \"actual\": [], \"expected\": [], \"passed\": true}], \"passed\": false}\n"},"broken":{"elapsed_ms":26.88,"exit_code":1,"observations":[{"actual":[4,6],"check":"equal shape","expected":[4,6],"passed":true},{"actual":[4],"check":"mismatch rejected","expected":null,"passed":false},{"actual":[],"check":"empty","expected":[],"passed":true},{"actual":[],"check":"singleton empty vector","expected":[],"passed":true},{"actual":[1],"check":"singleton values","expected":[1],"passed":true},{"actual":[],"check":"all empty shapes","expected":[],"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"equal shape\", \"actual\": [4, 6], \"expected\": [4, 6], \"passed\": true}, {\"check\": \"mismatch rejected\", \"actual\": [4], \"expected\": null, \"passed\": false}, {\"check\": \"empty\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton empty vector\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"singleton values\", \"actual\": [1], \"expected\": [1], \"passed\": true}, {\"check\": \"all empty shapes\", \"actual\": [], \"expected\": [], \"passed\": true}], \"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."}}