{"abstract":"Expert outputs are added without their gate weights.","category":"Mixture of experts dispatch","checks":6,"contract":"For [expert,weight,value] routes, return the scalar weighted sum.","contract_signature":"routes","evaluation_group":"moe-weighted-combine","failed_approach":"Using only the first route weight applies one coefficient to all experts.","family":"w3-mixture-of-experts-dispatch-weighted-combine","id":"FA-100596","implementations":{"attempt":{"sha256":"b2e97ea5f5086481276f05d66a8c3d6290f2870570d1f7455e309696e9b47093","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(routes):\n    return sum(routes[0][1]*v for e,w,v in routes) if routes else 0\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nrows=[['a',0.25,N],['b',0.75,2*N]]\ncheck('weighted values',solve(rows),1.75*N)\ncheck('unit weights',solve([['x',1,N]]),N)\ncheck('zero weight',solve([['x',0,N]]),0)\ncheck('empty',solve([]),0)\ncheck('negative output',solve([['x',0.5,-2*N]]),-N)\ncheck('three routes',solve([['a',0.2,N],['b',0.3,N],['c',0.5,N]]),N)\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":"b325f9724035b9d60b904aaa349d414284bae32367969775502dc302e477880f","source":"\"\"\"Failure Map reference implementation. Python standard library only.\"\"\"\nimport json\n\nN = 1\nobservations = []\ndef solve(routes):\n    return sum(v for e,w,v in routes)\ndef check(label, actual, expected):\n    observations.append({\"check\": label, \"actual\": actual, \"expected\": expected, \"passed\": actual == expected})\nrows=[['a',0.25,N],['b',0.75,2*N]]\ncheck('weighted values',solve(rows),1.75*N)\ncheck('unit weights',solve([['x',1,N]]),N)\ncheck('zero weight',solve([['x',0,N]]),0)\ncheck('empty',solve([]),0)\ncheck('negative output',solve([['x',0.5,-2*N]]),-N)\ncheck('three routes',solve([['a',0.2,N],['b',0.3,N],['c',0.5,N]]),N)\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-weighted-combine","generated_at":"2026-09-30T16:48:37.194261+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":"The combine stage sums expert values as if every route had unit weight.","sha256":"f3e45f1780cca85e74d273fa5af5d1881d3790230105d0a6862fe436c444034a","title":"Expert outputs are added without their gate weights · 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":33.267,"exit_code":1,"observations":[{"actual":0.75,"check":"weighted values","expected":1.75,"passed":false},{"actual":1,"check":"unit weights","expected":1,"passed":true},{"actual":0,"check":"zero weight","expected":0,"passed":true},{"actual":0,"check":"empty","expected":0,"passed":true},{"actual":-1.0,"check":"negative output","expected":-1,"passed":true},{"actual":0.6000000000000001,"check":"three routes","expected":1,"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"weighted values\", \"actual\": 0.75, \"expected\": 1.75, \"passed\": false}, {\"check\": \"unit weights\", \"actual\": 1, \"expected\": 1, \"passed\": true}, {\"check\": \"zero weight\", \"actual\": 0, \"expected\": 0, \"passed\": true}, {\"check\": \"empty\", \"actual\": 0, \"expected\": 0, \"passed\": true}, {\"check\": \"negative output\", \"actual\": -1.0, \"expected\": -1, \"passed\": true}, {\"check\": \"three routes\", \"actual\": 0.6000000000000001, \"expected\": 1, \"passed\": false}], \"passed\": false}\n"},"broken":{"elapsed_ms":29.537,"exit_code":1,"observations":[{"actual":3,"check":"weighted values","expected":1.75,"passed":false},{"actual":1,"check":"unit weights","expected":1,"passed":true},{"actual":1,"check":"zero weight","expected":0,"passed":false},{"actual":0,"check":"empty","expected":0,"passed":true},{"actual":-2,"check":"negative output","expected":-1,"passed":false},{"actual":3,"check":"three routes","expected":1,"passed":false}],"passed":false,"stderr":"","stdout":"{\"observations\": [{\"check\": \"weighted values\", \"actual\": 3, \"expected\": 1.75, \"passed\": false}, {\"check\": \"unit weights\", \"actual\": 1, \"expected\": 1, \"passed\": true}, {\"check\": \"zero weight\", \"actual\": 1, \"expected\": 0, \"passed\": false}, {\"check\": \"empty\", \"actual\": 0, \"expected\": 0, \"passed\": true}, {\"check\": \"negative output\", \"actual\": -2, \"expected\": -1, \"passed\": false}, {\"check\": \"three routes\", \"actual\": 3, \"expected\": 1, \"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."}}