{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OAH7242MF3TWNFKVZHHK76F2BR","short_pith_number":"pith:OAH7242M","canonical_record":{"source":{"id":"2505.19949","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T13:15:26Z","cross_cats_sorted":[],"title_canon_sha256":"b21cd8a38746a99b20f8de1070053965e77abc5a0251af998803a9cf5c401169","abstract_canon_sha256":"643e1e074c528de7da8cb7457fe1fbff7285a24904d32ad03221607c92d84afb"},"schema_version":"1.0"},"canonical_sha256":"700ffd734c2ee7669555c9ceaff8ba0c4d5ed06c31b18945eff89ce40820cbf6","source":{"kind":"arxiv","id":"2505.19949","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19949","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19949v1","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19949","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"OAH7242MF3TW","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_16","alias_value":"OAH7242MF3TWNFKV","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_8","alias_value":"OAH7242M","created_at":"2026-07-05T11:09:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OAH7242MF3TWNFKVZHHK76F2BR","target":"record","payload":{"canonical_record":{"source":{"id":"2505.19949","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T13:15:26Z","cross_cats_sorted":[],"title_canon_sha256":"b21cd8a38746a99b20f8de1070053965e77abc5a0251af998803a9cf5c401169","abstract_canon_sha256":"643e1e074c528de7da8cb7457fe1fbff7285a24904d32ad03221607c92d84afb"},"schema_version":"1.0"},"canonical_sha256":"700ffd734c2ee7669555c9ceaff8ba0c4d5ed06c31b18945eff89ce40820cbf6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:50.309115Z","signature_b64":"5P77vDEZxJPKI0FreHhBcymWwPuPMCRlQq/SBjPK20SwD0bNWQMgKvC/Gied7HpZ6y4FMt4Mxp7kuTwsMo+ODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"700ffd734c2ee7669555c9ceaff8ba0c4d5ed06c31b18945eff89ce40820cbf6","last_reissued_at":"2026-07-05T11:09:50.308610Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:50.308610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.19949","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wg4DDqBJrBLXlHBEIUk6epKa9rcU94xejDoxm2TkOvHZxix9N2QmBCTENF+A7WFJlhlCs2BFv+K6Jmye6PfNAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T19:17:38.630998Z"},"content_sha256":"451d1ae569579a35e22517a25e3f2c488cb67337dbb968ad60911c26d9b1cad0","schema_version":"1.0","event_id":"sha256:451d1ae569579a35e22517a25e3f2c488cb67337dbb968ad60911c26d9b1cad0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OAH7242MF3TWNFKVZHHK76F2BR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Which Data Attributes Stimulate Math and Code Reasoning? An Investigation via Influence Functions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hanwen Xu, Qingyuan Tian, Siqi Kou, Zhijie Deng, Zihao Zeng","submitted_at":"2025-05-26T13:15:26Z","abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable reasoning capabilities in math and coding, often bolstered by post-training on the chain-of-thoughts (CoTs) generated by stronger models. However, existing strategies for curating such training data predominantly rely on heuristics, limiting generalizability and failing to capture subtleties underlying in data. To address these limitations, we leverage influence functions to systematically attribute LLMs' reasoning ability on math and coding to individual training examples, sequences, and tokens, enabling deeper insights into effective "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19949","kind":"arxiv","version":1},"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/2505.19949/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":"2026-07-05T11:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ebOg1/zgos3XIvHhPs0VqOvwBHt4Rl8SPwNjU2zAP5y78NaZjZ/+ea8Xta98EJj/qUASbBXQUyWJXK/xlrJtCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T19:17:38.631504Z"},"content_sha256":"122c0df07d64f24f10043b4b0a67b81dd2899fbeee0831fff9a080929ca7f0a7","schema_version":"1.0","event_id":"sha256:122c0df07d64f24f10043b4b0a67b81dd2899fbeee0831fff9a080929ca7f0a7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OAH7242MF3TWNFKVZHHK76F2BR/bundle.json","state_url":"https://pith.science/pith/OAH7242MF3TWNFKVZHHK76F2BR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OAH7242MF3TWNFKVZHHK76F2BR/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":"2026-08-08T19:17:38Z","links":{"resolver":"https://pith.science/pith/OAH7242MF3TWNFKVZHHK76F2BR","bundle":"https://pith.science/pith/OAH7242MF3TWNFKVZHHK76F2BR/bundle.json","state":"https://pith.science/pith/OAH7242MF3TWNFKVZHHK76F2BR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OAH7242MF3TWNFKVZHHK76F2BR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OAH7242MF3TWNFKVZHHK76F2BR","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":"643e1e074c528de7da8cb7457fe1fbff7285a24904d32ad03221607c92d84afb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T13:15:26Z","title_canon_sha256":"b21cd8a38746a99b20f8de1070053965e77abc5a0251af998803a9cf5c401169"},"schema_version":"1.0","source":{"id":"2505.19949","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19949","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19949v1","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19949","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"OAH7242MF3TW","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_16","alias_value":"OAH7242MF3TWNFKV","created_at":"2026-07-05T11:09:50Z"},{"alias_kind":"pith_short_8","alias_value":"OAH7242M","created_at":"2026-07-05T11:09:50Z"}],"graph_snapshots":[{"event_id":"sha256:122c0df07d64f24f10043b4b0a67b81dd2899fbeee0831fff9a080929ca7f0a7","target":"graph","created_at":"2026-07-05T11:09:50Z","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/2505.19949/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable reasoning capabilities in math and coding, often bolstered by post-training on the chain-of-thoughts (CoTs) generated by stronger models. However, existing strategies for curating such training data predominantly rely on heuristics, limiting generalizability and failing to capture subtleties underlying in data. To address these limitations, we leverage influence functions to systematically attribute LLMs' reasoning ability on math and coding to individual training examples, sequences, and tokens, enabling deeper insights into effective ","authors_text":"Hanwen Xu, Qingyuan Tian, Siqi Kou, Zhijie Deng, Zihao Zeng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T13:15:26Z","title":"Which Data Attributes Stimulate Math and Code Reasoning? An Investigation via Influence Functions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19949","kind":"arxiv","version":1},"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":"sha256:451d1ae569579a35e22517a25e3f2c488cb67337dbb968ad60911c26d9b1cad0","target":"record","created_at":"2026-07-05T11:09:50Z","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":"643e1e074c528de7da8cb7457fe1fbff7285a24904d32ad03221607c92d84afb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T13:15:26Z","title_canon_sha256":"b21cd8a38746a99b20f8de1070053965e77abc5a0251af998803a9cf5c401169"},"schema_version":"1.0","source":{"id":"2505.19949","kind":"arxiv","version":1}},"canonical_sha256":"700ffd734c2ee7669555c9ceaff8ba0c4d5ed06c31b18945eff89ce40820cbf6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"700ffd734c2ee7669555c9ceaff8ba0c4d5ed06c31b18945eff89ce40820cbf6","first_computed_at":"2026-07-05T11:09:50.308610Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:50.308610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5P77vDEZxJPKI0FreHhBcymWwPuPMCRlQq/SBjPK20SwD0bNWQMgKvC/Gied7HpZ6y4FMt4Mxp7kuTwsMo+ODA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:50.309115Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19949","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:451d1ae569579a35e22517a25e3f2c488cb67337dbb968ad60911c26d9b1cad0","sha256:122c0df07d64f24f10043b4b0a67b81dd2899fbeee0831fff9a080929ca7f0a7"],"state_sha256":"4d08ba89fb1efd29c08c0db9ac6e1f912472de26f26e02b1ad979a1bcd204ab4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rXdV6BVmKpaqN06MmYAH01olBEezkZioFnNKGKAH5KdueXM1FK6zwf8q9UtWnKW9X208jzg20+8gHE+MN117BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T19:17:38.635946Z","bundle_sha256":"5c5244cb8478a88089a3ae1276d2d8d1f4446927019679857f078a4fcd06f740"}}