{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HRMKPDEBIAGZADOFPUHBKIRL4X","short_pith_number":"pith:HRMKPDEB","schema_version":"1.0","canonical_sha256":"3c58a78c81400d900dc57d0e15222be5c881f1a626423dfdd588b8f421ed6403","source":{"kind":"arxiv","id":"2605.29303","version":1},"attestation_state":"computed","paper":{"title":"Entropy-KL Divergence-based Token Masking: A Novel Approach for Selective Fine-tuning of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Enhong Chen, Mingdi Sun, Qi Liu, Tong Xu, Yi Zheng, Yongyi He, Zhefeng Wang, Zhi Zheng","submitted_at":"2026-05-28T03:36:05Z","abstract_excerpt":"Supervised fine-tuning (SFT) followed by reinforcement learning (RL) has become a standard post-training paradigm for large language models. This paradigm provides a cold-start for RL exploration, avoiding the inefficiency of pure RL where on-policy sampling yields insufficient positive samples. However, in practice, existing approaches often use a small amount of data for SFT initialization compared to the RL phase, which can cause the model to fit the limited samples and shift away from its pre-trained distribution. This distribution shift impedes the model's ability to effectively explore d"},"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":"2605.29303","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T03:36:05Z","cross_cats_sorted":[],"title_canon_sha256":"00681de600578348ee5cb316c39034ce06676b6f22cb2393b75c6349445b9b8d","abstract_canon_sha256":"382c1b86837aabb6487998c9c44c0a1cb1e53baf4df86dbe41d759ac1ac1ee5b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:30.976735Z","signature_b64":"S0SvKeX0PIFc7iOaUm8OcTuQXkHKiLapRlW9P6A3/I+/Oa987bX2VnYR5dLNUnVisllMnexPl44IHNDBx/7RCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c58a78c81400d900dc57d0e15222be5c881f1a626423dfdd588b8f421ed6403","last_reissued_at":"2026-05-29T01:05:30.976162Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:30.976162Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Entropy-KL Divergence-based Token Masking: A Novel Approach for Selective Fine-tuning of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Enhong Chen, Mingdi Sun, Qi Liu, Tong Xu, Yi Zheng, Yongyi He, Zhefeng Wang, Zhi Zheng","submitted_at":"2026-05-28T03:36:05Z","abstract_excerpt":"Supervised fine-tuning (SFT) followed by reinforcement learning (RL) has become a standard post-training paradigm for large language models. This paradigm provides a cold-start for RL exploration, avoiding the inefficiency of pure RL where on-policy sampling yields insufficient positive samples. However, in practice, existing approaches often use a small amount of data for SFT initialization compared to the RL phase, which can cause the model to fit the limited samples and shift away from its pre-trained distribution. This distribution shift impedes the model's ability to effectively explore d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29303","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/2605.29303/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":"2605.29303","created_at":"2026-05-29T01:05:30.976276+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29303v1","created_at":"2026-05-29T01:05:30.976276+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29303","created_at":"2026-05-29T01:05:30.976276+00:00"},{"alias_kind":"pith_short_12","alias_value":"HRMKPDEBIAGZ","created_at":"2026-05-29T01:05:30.976276+00:00"},{"alias_kind":"pith_short_16","alias_value":"HRMKPDEBIAGZADOF","created_at":"2026-05-29T01:05:30.976276+00:00"},{"alias_kind":"pith_short_8","alias_value":"HRMKPDEB","created_at":"2026-05-29T01:05:30.976276+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/HRMKPDEBIAGZADOFPUHBKIRL4X","json":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X.json","graph_json":"https://pith.science/api/pith-number/HRMKPDEBIAGZADOFPUHBKIRL4X/graph.json","events_json":"https://pith.science/api/pith-number/HRMKPDEBIAGZADOFPUHBKIRL4X/events.json","paper":"https://pith.science/paper/HRMKPDEB"},"agent_actions":{"view_html":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X","download_json":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X.json","view_paper":"https://pith.science/paper/HRMKPDEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29303&json=true","fetch_graph":"https://pith.science/api/pith-number/HRMKPDEBIAGZADOFPUHBKIRL4X/graph.json","fetch_events":"https://pith.science/api/pith-number/HRMKPDEBIAGZADOFPUHBKIRL4X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X/action/storage_attestation","attest_author":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X/action/author_attestation","sign_citation":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X/action/citation_signature","submit_replication":"https://pith.science/pith/HRMKPDEBIAGZADOFPUHBKIRL4X/action/replication_record"}},"created_at":"2026-05-29T01:05:30.976276+00:00","updated_at":"2026-05-29T01:05:30.976276+00:00"}