{"abstract":"Target gather uses the context position instead of its successor.","category":"Autoregressive target shifting","checks":6,"contract":"Contexts are [absolute_position,input_token] rows and targets maps absolute positions to tokens. Return [input,target_at_position_plus_one] for each context row.","contract_signature":"contexts, targets","evaluation_group":"w3-ars-position_supervision","failed_approach":"Using the enumeration offset as a target key fails when the context stream starts away from zero.","family":"w3-autoregressive-target-shifting-position-target-gather","id":"FA-98306","implementations":{"attempt":{"sha256":"bc75986eccd428f68867ea7d7a351ceb986a2aadbd2b84e2fc55d4f0cc44863f","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(contexts, targets):\n    return [[token,targets.get(i+1,targets.get(position,-1))] for i,(position,token) in enumerate(contexts)]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('shifted lookup',solve([[4,10],[5,11]],{4:90,5:12,6:13}),[[10,12],[11,13]])\ncheck('nonzero origin',solve([[N,7]],{N:7,N+1:8}),[[7,8]])\ncheck('negative positions',solve([[-2,1],[-1,2]],{-2:1,-1:2,0:3}),[[1,2],[2,3]])\ncheck('empty contexts',solve([],{1:8}),[])\ncheck('single context',solve([[20,4]],{20:4,21:9}),[[4,9]])\ncheck('target differs from input',solve([[0,2]],{0:2,1:8}),[[2,8]])\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":"630494ba666db7af7ad3592a6db7b36a65b3a5bf497f248eb2394bcabf1e1c0d","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(contexts, targets):\n    return [[token,targets[position]] for position,token in contexts]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\ncheck('shifted lookup',solve([[4,10],[5,11]],{4:90,5:12,6:13}),[[10,12],[11,13]])\ncheck('nonzero origin',solve([[N,7]],{N:7,N+1:8}),[[7,8]])\ncheck('negative positions',solve([[-2,1],[-1,2]],{-2:1,-1:2,0:3}),[[1,2],[2,3]])\ncheck('empty contexts',solve([],{1:8}),[])\ncheck('single context',solve([[20,4]],{20:4,21:9}),[[4,9]])\ncheck('target differs from input',solve([[0,2]],{0:2,1:8}),[[2,8]])\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 toy integer token sequences; this contract says nothing about model quality, batching, sampling, or production training frameworks. 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-autoregressive-target-shifting-position-target-gather","generated_at":"2026-09-30T16:48:11.500424+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A small deterministic token-stream contract makes next-token alignment errors inspectable.","root_cause":"A lookup reads the target stored at the context position rather than the next absolute position.","sha256":"009f4b74d6d757080f91a4396aabc978e4b8ca610201b0ed92e7c4a8aae004a0","title":"Target gather uses the context position instead of its successor · 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":28.467,"exit_code":1,"observations":[{"actual":[[10,90],[11,12]],"check":"shifted lookup","expected":[[10,12],[11,13]],"passed":false},{"actual":[[7,7]],"check":"nonzero origin","expected":[[7,8]],"passed":false},{"actual":[[1,1],[2,2]],"check":"negative positions","expected":[[1,2],[2,3]],"passed":false},{"actual":[],"check":"empty contexts","expected":[],"passed":true},{"actual":[[4,4]],"check":"single context","expected":[[4,9]],"passed":false},{"actual":[[2,8]],"check":"target differs from input","expected":[[2,8]],"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"shifted lookup\", \"actual\": [[10, 90], [11, 12]], \"expected\": [[10, 12], [11, 13]], \"passed\": false}, {\"check\": \"nonzero origin\", \"actual\": [[7, 7]], \"expected\": [[7, 8]], \"passed\": false}, {\"check\": \"negative positions\", \"actual\": [[1, 1], [2, 2]], \"expected\": [[1, 2], [2, 3]], \"passed\": false}, {\"check\": \"empty contexts\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"single context\", \"actual\": [[4, 4]], \"expected\": [[4, 9]], \"passed\": false}, {\"check\": \"target differs from input\", \"actual\": [[2, 8]], \"expected\": [[2, 8]], \"passed\": true}], \"passed\": false}\n"},"broken":{"elapsed_ms":27.2,"exit_code":1,"observations":[{"actual":[[10,90],[11,12]],"check":"shifted lookup","expected":[[10,12],[11,13]],"passed":false},{"actual":[[7,7]],"check":"nonzero origin","expected":[[7,8]],"passed":false},{"actual":[[1,1],[2,2]],"check":"negative positions","expected":[[1,2],[2,3]],"passed":false},{"actual":[],"check":"empty contexts","expected":[],"passed":true},{"actual":[[4,4]],"check":"single context","expected":[[4,9]],"passed":false},{"actual":[[2,2]],"check":"target differs from input","expected":[[2,8]],"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"shifted lookup\", \"actual\": [[10, 90], [11, 12]], \"expected\": [[10, 12], [11, 13]], \"passed\": false}, {\"check\": \"nonzero origin\", \"actual\": [[7, 7]], \"expected\": [[7, 8]], \"passed\": false}, {\"check\": \"negative positions\", \"actual\": [[1, 1], [2, 2]], \"expected\": [[1, 2], [2, 3]], \"passed\": false}, {\"check\": \"empty contexts\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"single context\", \"actual\": [[4, 4]], \"expected\": [[4, 9]], \"passed\": false}, {\"check\": \"target differs from input\", \"actual\": [[2, 2]], \"expected\": [[2, 8]], \"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."}}