{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:PC3MYR35WXXTKCGCSHXP54G4FA","short_pith_number":"pith:PC3MYR35","canonical_record":{"source":{"id":"2406.04197","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:55:53Z","cross_cats_sorted":[],"title_canon_sha256":"8cae6d495ed0c09c19ab3f137ff1640e7c0181a49e6c7aa90d3ff6205cf2da62","abstract_canon_sha256":"f33fe552de1320c5d98ec44803d0a6465a6a53d3071cc1f246cc5e0e07dd2d1a"},"schema_version":"1.0"},"canonical_sha256":"78b6cc477db5ef3508c291eefef0dc280a3b239810fee7a11d343bd93f6b9ec0","source":{"kind":"arxiv","id":"2406.04197","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04197","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04197v2","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04197","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_12","alias_value":"PC3MYR35WXXT","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_16","alias_value":"PC3MYR35WXXTKCGC","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_8","alias_value":"PC3MYR35","created_at":"2026-07-05T09:10:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:PC3MYR35WXXTKCGCSHXP54G4FA","target":"record","payload":{"canonical_record":{"source":{"id":"2406.04197","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:55:53Z","cross_cats_sorted":[],"title_canon_sha256":"8cae6d495ed0c09c19ab3f137ff1640e7c0181a49e6c7aa90d3ff6205cf2da62","abstract_canon_sha256":"f33fe552de1320c5d98ec44803d0a6465a6a53d3071cc1f246cc5e0e07dd2d1a"},"schema_version":"1.0"},"canonical_sha256":"78b6cc477db5ef3508c291eefef0dc280a3b239810fee7a11d343bd93f6b9ec0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:10:00.319963Z","signature_b64":"XH7w92JaMzLT+5nnWrWcaYZIzbyxcAynTl2knuP8olc/COTrUUXs+1HFDwW8PHjb/i0iKIDwuA5QXufW794YCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78b6cc477db5ef3508c291eefef0dc280a3b239810fee7a11d343bd93f6b9ec0","last_reissued_at":"2026-07-05T09:10:00.319465Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:10:00.319465Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.04197","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":"2026-07-05T09:10:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CTaMZNBOPk8f7J4s3SMr7z3oN3hhuQxLGec5ZL3OPGEBzqZLj1mjx7K9S5igsx/C827dxXrD08Cq9I+uP7k8AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:06:13.434453Z"},"content_sha256":"556816c82e23b1210d9ecd61ae0f71fafc06c66e05a4c53f3f80fe7cff9de0b0","schema_version":"1.0","event_id":"sha256:556816c82e23b1210d9ecd61ae0f71fafc06c66e05a4c53f3f80fe7cff9de0b0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:PC3MYR35WXXTKCGCSHXP54G4FA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DICE: Detecting In-distribution Contamination in LLM's Fine-tuning Phase for Math Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Juanzi Li, Kejian Zhu, Lei Hou, Shangqing Tu, Yushi Bai, Zijun Yao","submitted_at":"2024-06-06T15:55:53Z","abstract_excerpt":"The advancement of large language models (LLMs) relies on evaluation using public benchmarks, but data contamination can lead to overestimated performance. Previous researches focus on detecting contamination by determining whether the model has seen the exact same data during training. Besides, prior work has already shown that even training on data similar to benchmark data inflates performance, namely \\emph{In-distribution contamination}. In this work, we argue that in-distribution contamination can lead to the performance drop on OOD benchmarks. To effectively detect in-distribution contam"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04197","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/2406.04197/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":"2026-07-05T09:10:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VELf/wMBTOxMzrAltgZqIJ5uFecNwkB8kgmkI9pMMVY6r1h2Q6V9oltvKq4h1bJQSZKMSnaN7H2XSf7KTToHAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:06:13.434823Z"},"content_sha256":"76b9b58650544bd2aa2c71a6433f3976b7233eb78405fd8a2041b305d213c01e","schema_version":"1.0","event_id":"sha256:76b9b58650544bd2aa2c71a6433f3976b7233eb78405fd8a2041b305d213c01e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PC3MYR35WXXTKCGCSHXP54G4FA/bundle.json","state_url":"https://pith.science/pith/PC3MYR35WXXTKCGCSHXP54G4FA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PC3MYR35WXXTKCGCSHXP54G4FA/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":"2026-07-06T15:06:13Z","links":{"resolver":"https://pith.science/pith/PC3MYR35WXXTKCGCSHXP54G4FA","bundle":"https://pith.science/pith/PC3MYR35WXXTKCGCSHXP54G4FA/bundle.json","state":"https://pith.science/pith/PC3MYR35WXXTKCGCSHXP54G4FA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PC3MYR35WXXTKCGCSHXP54G4FA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PC3MYR35WXXTKCGCSHXP54G4FA","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":"f33fe552de1320c5d98ec44803d0a6465a6a53d3071cc1f246cc5e0e07dd2d1a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:55:53Z","title_canon_sha256":"8cae6d495ed0c09c19ab3f137ff1640e7c0181a49e6c7aa90d3ff6205cf2da62"},"schema_version":"1.0","source":{"id":"2406.04197","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04197","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04197v2","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04197","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_12","alias_value":"PC3MYR35WXXT","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_16","alias_value":"PC3MYR35WXXTKCGC","created_at":"2026-07-05T09:10:00Z"},{"alias_kind":"pith_short_8","alias_value":"PC3MYR35","created_at":"2026-07-05T09:10:00Z"}],"graph_snapshots":[{"event_id":"sha256:76b9b58650544bd2aa2c71a6433f3976b7233eb78405fd8a2041b305d213c01e","target":"graph","created_at":"2026-07-05T09:10:00Z","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/2406.04197/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The advancement of large language models (LLMs) relies on evaluation using public benchmarks, but data contamination can lead to overestimated performance. Previous researches focus on detecting contamination by determining whether the model has seen the exact same data during training. Besides, prior work has already shown that even training on data similar to benchmark data inflates performance, namely \\emph{In-distribution contamination}. In this work, we argue that in-distribution contamination can lead to the performance drop on OOD benchmarks. To effectively detect in-distribution contam","authors_text":"Juanzi Li, Kejian Zhu, Lei Hou, Shangqing Tu, Yushi Bai, Zijun Yao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:55:53Z","title":"DICE: Detecting In-distribution Contamination in LLM's Fine-tuning Phase for Math Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04197","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":"sha256:556816c82e23b1210d9ecd61ae0f71fafc06c66e05a4c53f3f80fe7cff9de0b0","target":"record","created_at":"2026-07-05T09:10:00Z","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":"f33fe552de1320c5d98ec44803d0a6465a6a53d3071cc1f246cc5e0e07dd2d1a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:55:53Z","title_canon_sha256":"8cae6d495ed0c09c19ab3f137ff1640e7c0181a49e6c7aa90d3ff6205cf2da62"},"schema_version":"1.0","source":{"id":"2406.04197","kind":"arxiv","version":2}},"canonical_sha256":"78b6cc477db5ef3508c291eefef0dc280a3b239810fee7a11d343bd93f6b9ec0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"78b6cc477db5ef3508c291eefef0dc280a3b239810fee7a11d343bd93f6b9ec0","first_computed_at":"2026-07-05T09:10:00.319465Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:10:00.319465Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XH7w92JaMzLT+5nnWrWcaYZIzbyxcAynTl2knuP8olc/COTrUUXs+1HFDwW8PHjb/i0iKIDwuA5QXufW794YCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:10:00.319963Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.04197","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:556816c82e23b1210d9ecd61ae0f71fafc06c66e05a4c53f3f80fe7cff9de0b0","sha256:76b9b58650544bd2aa2c71a6433f3976b7233eb78405fd8a2041b305d213c01e"],"state_sha256":"22b8c2df935450f8f47ca8f04fb142e838c261a7080b41cda35c2f88f6997101"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E7Qn+K4c0wtfFAWxjuQQDj0LoYEz3O+cbKCbwMnF6sGWISnfUAxDz7VZ69WcTzYs1GPP+ESs7HYco81d8e6WAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T15:06:13.436737Z","bundle_sha256":"c06b13adcf5e2d1362f025c123b9bee9ca4a5df1e5e0f3c917530592e95a6739"}}