{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Q6S3N2LF3MRY37HTX5NAFT3ELF","short_pith_number":"pith:Q6S3N2LF","schema_version":"1.0","canonical_sha256":"87a5b6e965db238dfcf3bf5a02cf64595ac82c2efbe32b88e5ead3d2eefe7250","source":{"kind":"arxiv","id":"2506.07833","version":2},"attestation_state":"computed","paper":{"title":"Improving Large Language Models with Concept-Aware Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Dacheng Tao, Jiaxing Huang, Michael K. Chen, Xikun Zhang","submitted_at":"2025-06-09T14:55:00Z","abstract_excerpt":"Large language models (LLMs) have become the cornerstone of modern AI. However, the existing paradigm of next-token prediction fundamentally limits their ability to form coherent, high-level concepts, making it a critical barrier to human-like understanding and reasoning. Take the phrase \"ribonucleic acid\" as an example: an LLM will first decompose it into tokens, i.e., artificial text fragments (\"rib\", \"on\", ...), then learn each token sequentially, rather than grasping the phrase as a unified, coherent semantic entity. This fragmented representation hinders deeper conceptual understanding an"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2506.07833","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T14:55:00Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ae2f1de4650ffa734c7cddc612c4e21c48559d86fa3f0b1356e871aedf9cf9d0","abstract_canon_sha256":"00fe4fb4947383773c316fbcc4b49da21aa9b53dcdd02ec4ad166e45fc390258"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:55.858555Z","signature_b64":"0nvm938tUQqU1WF5bdtRghlkYcmnRtRKJGjs74lvkmd6bHVzOGpq1AkjfoQtBHyFluXCKvgEdXQ1XS7zPMWbCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87a5b6e965db238dfcf3bf5a02cf64595ac82c2efbe32b88e5ead3d2eefe7250","last_reissued_at":"2026-07-05T11:20:55.857920Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:55.857920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Large Language Models with Concept-Aware Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Dacheng Tao, Jiaxing Huang, Michael K. Chen, Xikun Zhang","submitted_at":"2025-06-09T14:55:00Z","abstract_excerpt":"Large language models (LLMs) have become the cornerstone of modern AI. However, the existing paradigm of next-token prediction fundamentally limits their ability to form coherent, high-level concepts, making it a critical barrier to human-like understanding and reasoning. Take the phrase \"ribonucleic acid\" as an example: an LLM will first decompose it into tokens, i.e., artificial text fragments (\"rib\", \"on\", ...), then learn each token sequentially, rather than grasping the phrase as a unified, coherent semantic entity. This fragmented representation hinders deeper conceptual understanding an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.07833","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/2506.07833/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.07833","created_at":"2026-07-05T11:20:55.858008+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.07833v2","created_at":"2026-07-05T11:20:55.858008+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.07833","created_at":"2026-07-05T11:20:55.858008+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q6S3N2LF3MRY","created_at":"2026-07-05T11:20:55.858008+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q6S3N2LF3MRY37HT","created_at":"2026-07-05T11:20:55.858008+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q6S3N2LF","created_at":"2026-07-05T11:20:55.858008+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF","json":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF.json","graph_json":"https://pith.science/api/pith-number/Q6S3N2LF3MRY37HTX5NAFT3ELF/graph.json","events_json":"https://pith.science/api/pith-number/Q6S3N2LF3MRY37HTX5NAFT3ELF/events.json","paper":"https://pith.science/paper/Q6S3N2LF"},"agent_actions":{"view_html":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF","download_json":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF.json","view_paper":"https://pith.science/paper/Q6S3N2LF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.07833&json=true","fetch_graph":"https://pith.science/api/pith-number/Q6S3N2LF3MRY37HTX5NAFT3ELF/graph.json","fetch_events":"https://pith.science/api/pith-number/Q6S3N2LF3MRY37HTX5NAFT3ELF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF/action/storage_attestation","attest_author":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF/action/author_attestation","sign_citation":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF/action/citation_signature","submit_replication":"https://pith.science/pith/Q6S3N2LF3MRY37HTX5NAFT3ELF/action/replication_record"}},"created_at":"2026-07-05T11:20:55.858008+00:00","updated_at":"2026-07-05T11:20:55.858008+00:00"}