{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:H7FA6U3L7K3UA2AFIJA7SQ6F2B","short_pith_number":"pith:H7FA6U3L","canonical_record":{"source":{"id":"2502.14050","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-19T19:12:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2ffdcafda23af77efcf4f2bec5be1b59cbbc23cae81e893efd6c0bd8b0c990f2","abstract_canon_sha256":"3091d2b73028b13663e1efd0a874fbb17566e37fca6a09d4665f89b77d54e18a"},"schema_version":"1.0"},"canonical_sha256":"3fca0f536bfab74068054241f943c5d044fee2f87d6fc1760037ef392ab317c1","source":{"kind":"arxiv","id":"2502.14050","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14050","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14050v2","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14050","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_12","alias_value":"H7FA6U3L7K3U","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_16","alias_value":"H7FA6U3L7K3UA2AF","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_8","alias_value":"H7FA6U3L","created_at":"2026-07-05T10:42:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:H7FA6U3L7K3UA2AFIJA7SQ6F2B","target":"record","payload":{"canonical_record":{"source":{"id":"2502.14050","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-19T19:12:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2ffdcafda23af77efcf4f2bec5be1b59cbbc23cae81e893efd6c0bd8b0c990f2","abstract_canon_sha256":"3091d2b73028b13663e1efd0a874fbb17566e37fca6a09d4665f89b77d54e18a"},"schema_version":"1.0"},"canonical_sha256":"3fca0f536bfab74068054241f943c5d044fee2f87d6fc1760037ef392ab317c1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:15.053800Z","signature_b64":"0+Q/L7gPDK1F4xCyNXahHm23ZYp+lQFEpNgYot2s2ezUYSUeS5pxJ4TR9SrSlWIeqjXuP/7AjfWKCOxZLvzQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3fca0f536bfab74068054241f943c5d044fee2f87d6fc1760037ef392ab317c1","last_reissued_at":"2026-07-05T10:42:15.053302Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:15.053302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.14050","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":"2026-07-05T10:42:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0614xg1QW3+WzumGwriyDSAU0luqbOn2BHf0ORUAkvRWoRe4FbIelCuBw16rorS4B809Khh8djeLb/u25NoACQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:29:14.142922Z"},"content_sha256":"308d23ce7593c47d1bb1be28eb0e4d02478c357f88ba0bec2b453bbff7a8ee54","schema_version":"1.0","event_id":"sha256:308d23ce7593c47d1bb1be28eb0e4d02478c357f88ba0bec2b453bbff7a8ee54"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:H7FA6U3L7K3UA2AFIJA7SQ6F2B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Diversity-driven Data Selection for Language Model Tuning through Sparse Autoencoder","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Lijuan Liu, Madian Khabsa, Rui Hou, Shaoliang Nie, Suchin Gururangan, Ujjwal Karn, Xianjun Yang, Yuning Mao","submitted_at":"2025-02-19T19:12:34Z","abstract_excerpt":"Instruction tuning data are often quantity-saturated due to the large volume of data collection and fast model iteration, leaving data selection important but underexplored. Existing quality-driven data selection methods, such as LIMA (NeurIPS 2023 \\citep{zhou2024lima}) and AlpaGasus (ICLR 2024 \\citep{chenalpagasus}) generally ignore the equal importance of data diversity and complexity. In this work, we aim to design a diversity-aware data selection strategy and creatively propose using sparse autoencoders (SAEs) to tackle the challenge of data diversity measure. In addition, SAEs can also pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14050","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/2502.14050/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":"2026-07-05T10:42:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vnAiMS1W8X4+cqltGyPl7Iul4AJJVV9BjPqAedL7OKfwhP21Gk6ux2JBaNwqiDaYmyJ9brURBDXKpsRcnxSUBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T03:29:14.143426Z"},"content_sha256":"bc0bf4a623719c2716b4348e8862980333a795433ba32cad62eecd322ddd85f9","schema_version":"1.0","event_id":"sha256:bc0bf4a623719c2716b4348e8862980333a795433ba32cad62eecd322ddd85f9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/bundle.json","state_url":"https://pith.science/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/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":"2026-08-20T03:29:14Z","links":{"resolver":"https://pith.science/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B","bundle":"https://pith.science/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/bundle.json","state":"https://pith.science/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/H7FA6U3L7K3UA2AFIJA7SQ6F2B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:H7FA6U3L7K3UA2AFIJA7SQ6F2B","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":"3091d2b73028b13663e1efd0a874fbb17566e37fca6a09d4665f89b77d54e18a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-19T19:12:34Z","title_canon_sha256":"2ffdcafda23af77efcf4f2bec5be1b59cbbc23cae81e893efd6c0bd8b0c990f2"},"schema_version":"1.0","source":{"id":"2502.14050","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14050","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14050v2","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14050","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_12","alias_value":"H7FA6U3L7K3U","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_16","alias_value":"H7FA6U3L7K3UA2AF","created_at":"2026-07-05T10:42:15Z"},{"alias_kind":"pith_short_8","alias_value":"H7FA6U3L","created_at":"2026-07-05T10:42:15Z"}],"graph_snapshots":[{"event_id":"sha256:bc0bf4a623719c2716b4348e8862980333a795433ba32cad62eecd322ddd85f9","target":"graph","created_at":"2026-07-05T10:42:15Z","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/2502.14050/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction tuning data are often quantity-saturated due to the large volume of data collection and fast model iteration, leaving data selection important but underexplored. Existing quality-driven data selection methods, such as LIMA (NeurIPS 2023 \\citep{zhou2024lima}) and AlpaGasus (ICLR 2024 \\citep{chenalpagasus}) generally ignore the equal importance of data diversity and complexity. In this work, we aim to design a diversity-aware data selection strategy and creatively propose using sparse autoencoders (SAEs) to tackle the challenge of data diversity measure. In addition, SAEs can also pr","authors_text":"Lijuan Liu, Madian Khabsa, Rui Hou, Shaoliang Nie, Suchin Gururangan, Ujjwal Karn, Xianjun Yang, Yuning Mao","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-19T19:12:34Z","title":"Diversity-driven Data Selection for Language Model Tuning through Sparse Autoencoder"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14050","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":"sha256:308d23ce7593c47d1bb1be28eb0e4d02478c357f88ba0bec2b453bbff7a8ee54","target":"record","created_at":"2026-07-05T10:42:15Z","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":"3091d2b73028b13663e1efd0a874fbb17566e37fca6a09d4665f89b77d54e18a","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-19T19:12:34Z","title_canon_sha256":"2ffdcafda23af77efcf4f2bec5be1b59cbbc23cae81e893efd6c0bd8b0c990f2"},"schema_version":"1.0","source":{"id":"2502.14050","kind":"arxiv","version":2}},"canonical_sha256":"3fca0f536bfab74068054241f943c5d044fee2f87d6fc1760037ef392ab317c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3fca0f536bfab74068054241f943c5d044fee2f87d6fc1760037ef392ab317c1","first_computed_at":"2026-07-05T10:42:15.053302Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:42:15.053302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0+Q/L7gPDK1F4xCyNXahHm23ZYp+lQFEpNgYot2s2ezUYSUeS5pxJ4TR9SrSlWIeqjXuP/7AjfWKCOxZLvzQAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:42:15.053800Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.14050","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:308d23ce7593c47d1bb1be28eb0e4d02478c357f88ba0bec2b453bbff7a8ee54","sha256:bc0bf4a623719c2716b4348e8862980333a795433ba32cad62eecd322ddd85f9"],"state_sha256":"47e58d2312c93f7ef831cea98f6f01c29ef0d7e8045eed9e47a7ec4226ea386e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rJ6PdkvapC74FzWF5EPVUaWDa0gvfy5xQ23SqKtuJXv+kgu/r4TuZkhgW3y9DGQQwHAbJ0G1QQOP5IDLoJsjAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T03:29:14.147259Z","bundle_sha256":"7914dc4080544ebb7abad83d7aa864600e393c3ecf9bfced37fb7169b87431da"}}