{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PN2E3Q3QTSNEF3ULVT4Z2LQRRH","short_pith_number":"pith:PN2E3Q3Q","schema_version":"1.0","canonical_sha256":"7b744dc3709c9a42ee8bacf99d2e1189c7c11f8f62885b96875791450b026c44","source":{"kind":"arxiv","id":"2502.15920","version":2},"attestation_state":"computed","paper":{"title":"Self-Taught Agentic Long Context Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Emad Barsoum, Jialian Wu, Jiang Liu, Jingbo Shang, Xiaodong Yu, Ximeng Sun, Yufan Zhuang, Yusheng Su, Ze Wang, Zicheng Liu","submitted_at":"2025-02-21T20:29:36Z","abstract_excerpt":"Answering complex, long-context questions remains a major challenge for large language models (LLMs) as it requires effective question clarifications and context retrieval. We propose Agentic Long-Context Understanding (AgenticLU), a framework designed to enhance an LLM's understanding of such queries by integrating targeted self-clarification with contextual grounding within an agentic workflow. At the core of AgenticLU is Chain-of-Clarifications (CoC), where models refine their understanding through self-generated clarification questions and corresponding contextual groundings. By scaling in"},"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":"2502.15920","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-21T20:29:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"54c2dc2ee79711aa9040abbb7ba1fa2f2d2f9936af69cca7a37bc6bcfc188c99","abstract_canon_sha256":"50eec6c2b6acdbf26250a667fb7c29ab5cb313a9da334cbffd97847a67ccbfd9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:53.201786Z","signature_b64":"0ZR7/X8nDpQHPH2hYNARBFVMmQR16VgbjmoiXTRSSLYVrduTrwaNCNuDG4grqq8f2kPsJElhXG72bS7mqeNVAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b744dc3709c9a42ee8bacf99d2e1189c7c11f8f62885b96875791450b026c44","last_reissued_at":"2026-07-05T11:10:53.201249Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:53.201249Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Taught Agentic Long Context Understanding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Emad Barsoum, Jialian Wu, Jiang Liu, Jingbo Shang, Xiaodong Yu, Ximeng Sun, Yufan Zhuang, Yusheng Su, Ze Wang, Zicheng Liu","submitted_at":"2025-02-21T20:29:36Z","abstract_excerpt":"Answering complex, long-context questions remains a major challenge for large language models (LLMs) as it requires effective question clarifications and context retrieval. We propose Agentic Long-Context Understanding (AgenticLU), a framework designed to enhance an LLM's understanding of such queries by integrating targeted self-clarification with contextual grounding within an agentic workflow. At the core of AgenticLU is Chain-of-Clarifications (CoC), where models refine their understanding through self-generated clarification questions and corresponding contextual groundings. By scaling in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.15920","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/2502.15920/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":"2502.15920","created_at":"2026-07-05T11:10:53.201324+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.15920v2","created_at":"2026-07-05T11:10:53.201324+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.15920","created_at":"2026-07-05T11:10:53.201324+00:00"},{"alias_kind":"pith_short_12","alias_value":"PN2E3Q3QTSNE","created_at":"2026-07-05T11:10:53.201324+00:00"},{"alias_kind":"pith_short_16","alias_value":"PN2E3Q3QTSNEF3UL","created_at":"2026-07-05T11:10:53.201324+00:00"},{"alias_kind":"pith_short_8","alias_value":"PN2E3Q3Q","created_at":"2026-07-05T11:10:53.201324+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.09725","citing_title":"Efficient Remote KV Cache Reuse with GPU-native Video Codec","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH","json":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH.json","graph_json":"https://pith.science/api/pith-number/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/graph.json","events_json":"https://pith.science/api/pith-number/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/events.json","paper":"https://pith.science/paper/PN2E3Q3Q"},"agent_actions":{"view_html":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH","download_json":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH.json","view_paper":"https://pith.science/paper/PN2E3Q3Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.15920&json=true","fetch_graph":"https://pith.science/api/pith-number/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/graph.json","fetch_events":"https://pith.science/api/pith-number/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/action/storage_attestation","attest_author":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/action/author_attestation","sign_citation":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/action/citation_signature","submit_replication":"https://pith.science/pith/PN2E3Q3QTSNEF3ULVT4Z2LQRRH/action/replication_record"}},"created_at":"2026-07-05T11:10:53.201324+00:00","updated_at":"2026-07-05T11:10:53.201324+00:00"}