{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:BOQKQ56GG2XIIMSWWWAMPQJAAY","short_pith_number":"pith:BOQKQ56G","canonical_record":{"source":{"id":"2407.12220","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-17T00:06:30Z","cross_cats_sorted":["cs.CL","cs.CY"],"title_canon_sha256":"8632f40fb3b79d78d6498a5d3311810963f1c1f872f1d616f3f95c6656b60abc","abstract_canon_sha256":"50052fe9ee36cb38dea8c72fe2925183d1a48388129af288d9baa896db5427d6"},"schema_version":"1.0"},"canonical_sha256":"0ba0a877c636ae843256b580c7c1200615beec14ae5880a5154ee173f1d00d65","source":{"kind":"arxiv","id":"2407.12220","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.12220","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"arxiv_version","alias_value":"2407.12220v2","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.12220","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_12","alias_value":"BOQKQ56GG2XI","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_16","alias_value":"BOQKQ56GG2XIIMSW","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_8","alias_value":"BOQKQ56G","created_at":"2026-07-05T09:28:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:BOQKQ56GG2XIIMSWWWAMPQJAAY","target":"record","payload":{"canonical_record":{"source":{"id":"2407.12220","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-17T00:06:30Z","cross_cats_sorted":["cs.CL","cs.CY"],"title_canon_sha256":"8632f40fb3b79d78d6498a5d3311810963f1c1f872f1d616f3f95c6656b60abc","abstract_canon_sha256":"50052fe9ee36cb38dea8c72fe2925183d1a48388129af288d9baa896db5427d6"},"schema_version":"1.0"},"canonical_sha256":"0ba0a877c636ae843256b580c7c1200615beec14ae5880a5154ee173f1d00d65","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:28:33.392845Z","signature_b64":"Mig+aU6N92IDz23vCKs8G5mqo31O0wAsT5QFKBV3ziUdvjgCIi34qb3FcgnxeDEGau0TOVGyeytqsfdwjUdoBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ba0a877c636ae843256b580c7c1200615beec14ae5880a5154ee173f1d00d65","last_reissued_at":"2026-07-05T09:28:33.392300Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:28:33.392300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.12220","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:28:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pLnXmpfb79QPe3j7vir2Vv6vi9L2fTEn9OKOCMv6b1vN94FKLxZI4LR2kMDggGLQjtIJJwMASpqO4oLuULIgDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T04:06:42.250699Z"},"content_sha256":"b63e50c1760611815e325b2797d70ce61c59287898e25a8b38c1b10bd96ac659","schema_version":"1.0","event_id":"sha256:b63e50c1760611815e325b2797d70ce61c59287898e25a8b38c1b10bd96ac659"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:BOQKQ56GG2XIIMSWWWAMPQJAAY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Questionable practices in machine learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY"],"primary_cat":"cs.LG","authors_text":"Gavin Leech, Juan J. Vazquez, Laurence Aitchison, Misha Yagudin, Niclas Kupper","submitted_at":"2024-07-17T00:06:30Z","abstract_excerpt":"Evaluating modern ML models is hard. The strong incentive for researchers and companies to report a state-of-the-art result on some metric often leads to questionable research practices (QRPs): bad practices which fall short of outright research fraud. We describe 44 such practices which can undermine reported results, giving examples where possible. Our list emphasises the evaluation of large language models (LLMs) on public benchmarks. We also discuss \"irreproducible research practices\", i.e. decisions that make it difficult or impossible for other researchers to reproduce, build on or audit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.12220","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2407.12220/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:28:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/stXQiSlzxAWQZbDx8jfLjLKywoupHknUX3ng3w45k7cPMaq6QAlmq3ab4RlsOudafhPxkAFnyJ4cINfgXmNDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T04:06:42.251397Z"},"content_sha256":"f05ef754e4d7ee6ad97a9a232316121c5d0826a687e8e7ef495d1c5381ab4317","schema_version":"1.0","event_id":"sha256:f05ef754e4d7ee6ad97a9a232316121c5d0826a687e8e7ef495d1c5381ab4317"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/bundle.json","state_url":"https://pith.science/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T04:06:42Z","links":{"resolver":"https://pith.science/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY","bundle":"https://pith.science/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/bundle.json","state":"https://pith.science/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BOQKQ56GG2XIIMSWWWAMPQJAAY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BOQKQ56GG2XIIMSWWWAMPQJAAY","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"50052fe9ee36cb38dea8c72fe2925183d1a48388129af288d9baa896db5427d6","cross_cats_sorted":["cs.CL","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-17T00:06:30Z","title_canon_sha256":"8632f40fb3b79d78d6498a5d3311810963f1c1f872f1d616f3f95c6656b60abc"},"schema_version":"1.0","source":{"id":"2407.12220","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.12220","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"arxiv_version","alias_value":"2407.12220v2","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.12220","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_12","alias_value":"BOQKQ56GG2XI","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_16","alias_value":"BOQKQ56GG2XIIMSW","created_at":"2026-07-05T09:28:33Z"},{"alias_kind":"pith_short_8","alias_value":"BOQKQ56G","created_at":"2026-07-05T09:28:33Z"}],"graph_snapshots":[{"event_id":"sha256:f05ef754e4d7ee6ad97a9a232316121c5d0826a687e8e7ef495d1c5381ab4317","target":"graph","created_at":"2026-07-05T09:28:33Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2407.12220/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating modern ML models is hard. The strong incentive for researchers and companies to report a state-of-the-art result on some metric often leads to questionable research practices (QRPs): bad practices which fall short of outright research fraud. We describe 44 such practices which can undermine reported results, giving examples where possible. Our list emphasises the evaluation of large language models (LLMs) on public benchmarks. We also discuss \"irreproducible research practices\", i.e. decisions that make it difficult or impossible for other researchers to reproduce, build on or audit","authors_text":"Gavin Leech, Juan J. Vazquez, Laurence Aitchison, Misha Yagudin, Niclas Kupper","cross_cats":["cs.CL","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-17T00:06:30Z","title":"Questionable practices in machine learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.12220","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b63e50c1760611815e325b2797d70ce61c59287898e25a8b38c1b10bd96ac659","target":"record","created_at":"2026-07-05T09:28:33Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"50052fe9ee36cb38dea8c72fe2925183d1a48388129af288d9baa896db5427d6","cross_cats_sorted":["cs.CL","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-17T00:06:30Z","title_canon_sha256":"8632f40fb3b79d78d6498a5d3311810963f1c1f872f1d616f3f95c6656b60abc"},"schema_version":"1.0","source":{"id":"2407.12220","kind":"arxiv","version":2}},"canonical_sha256":"0ba0a877c636ae843256b580c7c1200615beec14ae5880a5154ee173f1d00d65","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0ba0a877c636ae843256b580c7c1200615beec14ae5880a5154ee173f1d00d65","first_computed_at":"2026-07-05T09:28:33.392300Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:28:33.392300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Mig+aU6N92IDz23vCKs8G5mqo31O0wAsT5QFKBV3ziUdvjgCIi34qb3FcgnxeDEGau0TOVGyeytqsfdwjUdoBA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:28:33.392845Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.12220","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b63e50c1760611815e325b2797d70ce61c59287898e25a8b38c1b10bd96ac659","sha256:f05ef754e4d7ee6ad97a9a232316121c5d0826a687e8e7ef495d1c5381ab4317"],"state_sha256":"bc5b3d28ef1ad7cc905d73c82b6d005ab153b9234b0374b149b9153c8cddb5b6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ruyprsNnoLrR3r7C5UZkFL5UiRkybQVaqcNE2OB4lqSAn/3en7YLpuQknc4vb63NAiS3rRnWp435MZFAy+tpAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T04:06:42.256296Z","bundle_sha256":"3f0126a8535983e8ba3958be5fb8a00c841cdf4153d490daa5ee620d1c6c228f"}}