{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z4Y3T3MPIN7MMTJAXWMOWT4SB3","short_pith_number":"pith:Z4Y3T3MP","schema_version":"1.0","canonical_sha256":"cf31b9ed8f437ec64d20bd98eb4f920ed5b9dfe77a560c2ff94b9811da4c44fd","source":{"kind":"arxiv","id":"2504.21773","version":4},"attestation_state":"computed","paper":{"title":"MAC-Tuning: LLM Multi-Compositional Problem Reasoning with Enhanced Knowledge Boundary Awareness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Junsheng Huang, Qingyun Wang, Sandeep Polisetty, Yi.R Fung, Yucheng Huang, Zhitao He","submitted_at":"2025-04-30T16:17:53Z","abstract_excerpt":"The hallucination of non-existent facts by LLMs is an important problem given its widespread adoption across various applications. Previous research addresses this problem by analyzing the internal parameterized knowledge boundaries to estimate confidence. However, these studies focus on the single-problem setting and have not explored the more challenging multi-problem setting, which requires accurately answering multiple questions simultaneously. We introduce a novel method for the multi-problem setting, Multiple Answers and Confidence Stepwise Tuning (MAC-Tuning), that separates the learnin"},"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":"2504.21773","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-30T16:17:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0a7b5a71eeac379393a4956b7caa6dfc25aaab250303d149a8461f0f953b9884","abstract_canon_sha256":"50fb5ff88f9e19b566bed357929384cab7616cc77a13cef9f55f0ff1f9a161e9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:17.439477Z","signature_b64":"K667VH0bAlF0cd2AqEKzwgn/AmsF7/OQkcKf353HmDMAV5vBmLuADjp8MT24A2UNk2hbyR9Lo+BaqBz3ivPTBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf31b9ed8f437ec64d20bd98eb4f920ed5b9dfe77a560c2ff94b9811da4c44fd","last_reissued_at":"2026-07-05T12:11:17.438931Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:17.438931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAC-Tuning: LLM Multi-Compositional Problem Reasoning with Enhanced Knowledge Boundary Awareness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Junsheng Huang, Qingyun Wang, Sandeep Polisetty, Yi.R Fung, Yucheng Huang, Zhitao He","submitted_at":"2025-04-30T16:17:53Z","abstract_excerpt":"The hallucination of non-existent facts by LLMs is an important problem given its widespread adoption across various applications. Previous research addresses this problem by analyzing the internal parameterized knowledge boundaries to estimate confidence. However, these studies focus on the single-problem setting and have not explored the more challenging multi-problem setting, which requires accurately answering multiple questions simultaneously. We introduce a novel method for the multi-problem setting, Multiple Answers and Confidence Stepwise Tuning (MAC-Tuning), that separates the learnin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.21773","kind":"arxiv","version":4},"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/2504.21773/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":"2504.21773","created_at":"2026-07-05T12:11:17.438989+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.21773v4","created_at":"2026-07-05T12:11:17.438989+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.21773","created_at":"2026-07-05T12:11:17.438989+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z4Y3T3MPIN7M","created_at":"2026-07-05T12:11:17.438989+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z4Y3T3MPIN7MMTJA","created_at":"2026-07-05T12:11:17.438989+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z4Y3T3MP","created_at":"2026-07-05T12:11:17.438989+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/Z4Y3T3MPIN7MMTJAXWMOWT4SB3","json":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3.json","graph_json":"https://pith.science/api/pith-number/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/graph.json","events_json":"https://pith.science/api/pith-number/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/events.json","paper":"https://pith.science/paper/Z4Y3T3MP"},"agent_actions":{"view_html":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3","download_json":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3.json","view_paper":"https://pith.science/paper/Z4Y3T3MP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.21773&json=true","fetch_graph":"https://pith.science/api/pith-number/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/graph.json","fetch_events":"https://pith.science/api/pith-number/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/action/storage_attestation","attest_author":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/action/author_attestation","sign_citation":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/action/citation_signature","submit_replication":"https://pith.science/pith/Z4Y3T3MPIN7MMTJAXWMOWT4SB3/action/replication_record"}},"created_at":"2026-07-05T12:11:17.438989+00:00","updated_at":"2026-07-05T12:11:17.438989+00:00"}