{"abstract":"Low-confidence routes survive the gate threshold.","category":"Mixture of experts dispatch","checks":6,"contract":"Keep each [token,expert,score] route with score greater than or equal to threshold, preserving order.","contract_signature":"routes,threshold","evaluation_group":"xmed-live-window-prune-absolute-sequence","failed_approach":"Using a strict threshold also discards routes exactly on the allowed boundary.","family":"w3-mixture-of-experts-dispatch-gate-threshold","id":"FA-100576","implementations":{"attempt":{"sha256":"37b386f6d44a22286fd1a7544d16b79e65ad67ec877e91fafb260743cad91cf6","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(routes,threshold):\n    return [r for r in routes if r[2]>threshold]\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nrows=[[0,'a',N],[1,'b',N-1],[2,'c',N+1]]\ncheck('below removed',solve(rows,N),[rows[0],rows[2]])\ncheck('exact included',solve(rows,N-1),rows)\ncheck('above',solve(rows,N+2),[])\ncheck('empty',solve([],N),[])\ncheck('all pass',solve(rows,0),rows)\ncheck('negative boundary',solve([[0,'x',-1]],-1),[[0,'x',-1]])\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":"320a5b61bedaa8aee2ce2716d808698e989653fbfa927f5ff31f1b669d2dda10","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(routes,threshold):\n    return routes\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nrows=[[0,'a',N],[1,'b',N-1],[2,'c',N+1]]\ncheck('below removed',solve(rows,N),[rows[0],rows[2]])\ncheck('exact included',solve(rows,N-1),rows)\ncheck('above',solve(rows,N+2),[])\ncheck('empty',solve([],N),[])\ncheck('all pass',solve(rows,0),rows)\ncheck('negative boundary',solve([[0,'x',-1]],-1),[[0,'x',-1]])\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 finite toy scores and routes; the contract does not model a production model, accelerator, or learned routing quality. 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-mixture-of-experts-dispatch-gate-threshold","generated_at":"2026-09-30T16:48:37.194144+00:00","license":"CC0-1.0","python":"3.12.14","seed":1,"split":"open-access"},"relevance":"A deterministic toy router makes this dispatch decision independently inspectable.","root_cause":"Threshold filtering is applied only after capacity selection.","sha256":"3986138db861f8ee0c912e44b4412221102d0762b3032c478d8f86f628480d6c","title":"Low-confidence routes survive the gate threshold · 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":33.936,"exit_code":1,"observations":[{"actual":[[2,"c",2]],"check":"below removed","expected":[[0,"a",1],[2,"c",2]],"passed":false},{"actual":[[0,"a",1],[2,"c",2]],"check":"exact included","expected":[[0,"a",1],[1,"b",0],[2,"c",2]],"passed":false},{"actual":[],"check":"above","expected":[],"passed":true},{"actual":[],"check":"empty","expected":[],"passed":true},{"actual":[[0,"a",1],[2,"c",2]],"check":"all pass","expected":[[0,"a",1],[1,"b",0],[2,"c",2]],"passed":false},{"actual":[],"check":"negative boundary","expected":[[0,"x",-1]],"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"below removed\", \"actual\": [[2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [2, \"c\", 2]], \"passed\": false}, {\"check\": \"exact included\", \"actual\": [[0, \"a\", 1], [2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"passed\": false}, {\"check\": \"above\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"empty\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"all pass\", \"actual\": [[0, \"a\", 1], [2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"passed\": false}, {\"check\": \"negative boundary\", \"actual\": [], \"expected\": [[0, \"x\", -1]], \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":29.907,"exit_code":1,"observations":[{"actual":[[0,"a",1],[1,"b",0],[2,"c",2]],"check":"below removed","expected":[[0,"a",1],[2,"c",2]],"passed":false},{"actual":[[0,"a",1],[1,"b",0],[2,"c",2]],"check":"exact included","expected":[[0,"a",1],[1,"b",0],[2,"c",2]],"passed":true},{"actual":[[0,"a",1],[1,"b",0],[2,"c",2]],"check":"above","expected":[],"passed":false},{"actual":[],"check":"empty","expected":[],"passed":true},{"actual":[[0,"a",1],[1,"b",0],[2,"c",2]],"check":"all pass","expected":[[0,"a",1],[1,"b",0],[2,"c",2]],"passed":true},{"actual":[[0,"x",-1]],"check":"negative boundary","expected":[[0,"x",-1]],"passed":true}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"below removed\", \"actual\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [2, \"c\", 2]], \"passed\": false}, {\"check\": \"exact included\", \"actual\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"passed\": true}, {\"check\": \"above\", \"actual\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"expected\": [], \"passed\": false}, {\"check\": \"empty\", \"actual\": [], \"expected\": [], \"passed\": true}, {\"check\": \"all pass\", \"actual\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"expected\": [[0, \"a\", 1], [1, \"b\", 0], [2, \"c\", 2]], \"passed\": true}, {\"check\": \"negative boundary\", \"actual\": [[0, \"x\", -1]], \"expected\": [[0, \"x\", -1]], \"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."}}