{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:A67KMUY5PYDJGUIKI3EDESC6C5","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":"0d629756878a422d0cc689d1186cd990e0b08ae32e475fc4a9c3ec5969a03a6f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T16:05:10Z","title_canon_sha256":"8f4fd79196329fded6f75be299f48a44a741718c90a6e4154e8e9e8e193c1c6a"},"schema_version":"1.0","source":{"id":"2505.20161","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.20161","created_at":"2026-07-05T11:09:53Z"},{"alias_kind":"arxiv_version","alias_value":"2505.20161v1","created_at":"2026-07-05T11:09:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20161","created_at":"2026-07-05T11:09:53Z"},{"alias_kind":"pith_short_12","alias_value":"A67KMUY5PYDJ","created_at":"2026-07-05T11:09:53Z"},{"alias_kind":"pith_short_16","alias_value":"A67KMUY5PYDJGUIK","created_at":"2026-07-05T11:09:53Z"},{"alias_kind":"pith_short_8","alias_value":"A67KMUY5","created_at":"2026-07-05T11:09:53Z"}],"graph_snapshots":[{"event_id":"sha256:36bf804e89acec3a20c27c6f36f81c99e5be83105843e36d96e234ecf4bedb1f","target":"graph","created_at":"2026-07-05T11:09:53Z","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/2505.20161/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Effective generalization in language models depends critically on the diversity of their training data. Yet existing diversity metrics often fall short of this goal, relying on surface-level heuristics that are decoupled from model behavior. This motivates us to ask: What kind of diversity in training data actually drives generalization in language models -- and how can we measure and amplify it? Through large-scale empirical analyses spanning over 300 training runs, carefully controlled for data scale and quality, we show that data diversity can be a strong predictor of generalization in LLM ","authors_text":"Bryan Catanzaro, David Acuna, Jaehun Jung, Mohammad Shoeybi, Mostafa Patwary, Seungju Han, Shrimai Prabhumoye, Skyler Hallinan, Ximing Lu, Yejin Choi","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T16:05:10Z","title":"Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20161","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:a328b0d07f98f4d82541f13e188d4cfdef22354c0ccdf9b7eb86406b650f1210","target":"record","created_at":"2026-07-05T11:09:53Z","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":"0d629756878a422d0cc689d1186cd990e0b08ae32e475fc4a9c3ec5969a03a6f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-26T16:05:10Z","title_canon_sha256":"8f4fd79196329fded6f75be299f48a44a741718c90a6e4154e8e9e8e193c1c6a"},"schema_version":"1.0","source":{"id":"2505.20161","kind":"arxiv","version":1}},"canonical_sha256":"07bea6531d7e0693510a46c832485e1773aed0f9df63e00248a07af2c07741bb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"07bea6531d7e0693510a46c832485e1773aed0f9df63e00248a07af2c07741bb","first_computed_at":"2026-07-05T11:09:53.551231Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:53.551231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"K6nuyHM5UQQcwbL3Ji5iwgiin7witnQ8wTK9VeX+kVBIaZcbVA7oF6L407Dd2FRPxgP/BRpD/9b1ePyt5hc8Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:53.551729Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.20161","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a328b0d07f98f4d82541f13e188d4cfdef22354c0ccdf9b7eb86406b650f1210","sha256:36bf804e89acec3a20c27c6f36f81c99e5be83105843e36d96e234ecf4bedb1f"],"state_sha256":"15d64a279ef0335ab2103e9f34be8e6e4489ff4bdabc87b7782e563d0871bdcb"}