{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RK7NVID2KJRZQJXN26P3ZDL2VS","short_pith_number":"pith:RK7NVID2","schema_version":"1.0","canonical_sha256":"8abedaa07a52639826edd79fbc8d7aacb1ba8205b5d83f6090936ac2f1061c55","source":{"kind":"arxiv","id":"2411.06208","version":3},"attestation_state":"computed","paper":{"title":"IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cheng Fu, Fei Huang, Haiyang Yu, Xinghua Zhang, Yongbin Li","submitted_at":"2024-11-09T15:12:43Z","abstract_excerpt":"In the realm of large language models (LLMs), the ability of models to accurately follow instructions is paramount as more agents and applications leverage LLMs for construction, where the complexity of instructions are rapidly increasing. However, on the one hand, there is only a certain amount of complex instruction evaluation data; on the other hand, there are no dedicated algorithms to improve the ability to follow complex instructions. To this end, this paper introduces TRACE, a benchmark for improving and evaluating the complex instructionfollowing ability, which consists of 120K trainin"},"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":"2411.06208","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-09T15:12:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c20713400cbd972b8b452dadaf2d6aeeba2b71a6b36d94256af6398b0cc8f03d","abstract_canon_sha256":"999e429468dd66b60317798e59e1d51ccffb99cb24f8089f77e2e16d1dfcba8f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:25.352219Z","signature_b64":"zQKfnEtIFLbVwC7V+ekpKprgRBfaz02R7v9TvNKagnu4mmArlyiYa9ymVVikZZ0eBxFOlzV9sSbxcsiw0sHABg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8abedaa07a52639826edd79fbc8d7aacb1ba8205b5d83f6090936ac2f1061c55","last_reissued_at":"2026-07-05T11:38:25.351669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:25.351669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cheng Fu, Fei Huang, Haiyang Yu, Xinghua Zhang, Yongbin Li","submitted_at":"2024-11-09T15:12:43Z","abstract_excerpt":"In the realm of large language models (LLMs), the ability of models to accurately follow instructions is paramount as more agents and applications leverage LLMs for construction, where the complexity of instructions are rapidly increasing. However, on the one hand, there is only a certain amount of complex instruction evaluation data; on the other hand, there are no dedicated algorithms to improve the ability to follow complex instructions. To this end, this paper introduces TRACE, a benchmark for improving and evaluating the complex instructionfollowing ability, which consists of 120K trainin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.06208","kind":"arxiv","version":3},"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/2411.06208/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":"2411.06208","created_at":"2026-07-05T11:38:25.351743+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.06208v3","created_at":"2026-07-05T11:38:25.351743+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.06208","created_at":"2026-07-05T11:38:25.351743+00:00"},{"alias_kind":"pith_short_12","alias_value":"RK7NVID2KJRZ","created_at":"2026-07-05T11:38:25.351743+00:00"},{"alias_kind":"pith_short_16","alias_value":"RK7NVID2KJRZQJXN","created_at":"2026-07-05T11:38:25.351743+00:00"},{"alias_kind":"pith_short_8","alias_value":"RK7NVID2","created_at":"2026-07-05T11:38:25.351743+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.09942","citing_title":"VerIF: Verification Engineering for Reinforcement Learning in Instruction Following","ref_index":61,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS","json":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS.json","graph_json":"https://pith.science/api/pith-number/RK7NVID2KJRZQJXN26P3ZDL2VS/graph.json","events_json":"https://pith.science/api/pith-number/RK7NVID2KJRZQJXN26P3ZDL2VS/events.json","paper":"https://pith.science/paper/RK7NVID2"},"agent_actions":{"view_html":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS","download_json":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS.json","view_paper":"https://pith.science/paper/RK7NVID2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.06208&json=true","fetch_graph":"https://pith.science/api/pith-number/RK7NVID2KJRZQJXN26P3ZDL2VS/graph.json","fetch_events":"https://pith.science/api/pith-number/RK7NVID2KJRZQJXN26P3ZDL2VS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS/action/storage_attestation","attest_author":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS/action/author_attestation","sign_citation":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS/action/citation_signature","submit_replication":"https://pith.science/pith/RK7NVID2KJRZQJXN26P3ZDL2VS/action/replication_record"}},"created_at":"2026-07-05T11:38:25.351743+00:00","updated_at":"2026-07-05T11:38:25.351743+00:00"}