{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2X3FIIGXKDW7FUQQHN2EWIWWEU","short_pith_number":"pith:2X3FIIGX","schema_version":"1.0","canonical_sha256":"d5f65420d750edf2d2103b744b22d6252453b4e45aaca56124378da76bf48514","source":{"kind":"arxiv","id":"2410.09342","version":1},"attestation_state":"computed","paper":{"title":"LLM$\\times$MapReduce: Simplified Long-Sequence Processing using Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chong Li, Haoyu Wang, Maosong Sun, Qi Shi, Rongqiao An, Shuo Wang, Xiaodong Shi, Xinyi Chen, Xu Han, Yu Chao, Zhili Li, Zhixing Tan, Zhiyuan Liu, Zihan Zhou","submitted_at":"2024-10-12T03:13:44Z","abstract_excerpt":"Enlarging the context window of large language models (LLMs) has become a crucial research area, particularly for applications involving extremely long texts. In this work, we propose a novel training-free framework for processing long texts, utilizing a divide-and-conquer strategy to achieve comprehensive document understanding. The proposed LLM$\\times$MapReduce framework splits the entire document into several chunks for LLMs to read and then aggregates the intermediate answers to produce the final output. The main challenge for divide-and-conquer long text processing frameworks lies in the "},"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":"2410.09342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-12T03:13:44Z","cross_cats_sorted":[],"title_canon_sha256":"b195ff14c5777c8ab9dbf7574e5e65972e4f6678e95574cf41f94501e0364b8f","abstract_canon_sha256":"bfff07aa292ee5743b26e3b3f9d524a1a5726a7f8cc1a22a499d58b915ba1bef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:19:37.942047Z","signature_b64":"mvTv0Q9aSR2uL+jOxO5630vghYrtE8Z1bynXkBCWHq+FRnpZvAtd7HeH88yLGths7r8dHXr9k8WajQ+ucf92BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d5f65420d750edf2d2103b744b22d6252453b4e45aaca56124378da76bf48514","last_reissued_at":"2026-07-05T09:19:37.941620Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:19:37.941620Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM$\\times$MapReduce: Simplified Long-Sequence Processing using Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chong Li, Haoyu Wang, Maosong Sun, Qi Shi, Rongqiao An, Shuo Wang, Xiaodong Shi, Xinyi Chen, Xu Han, Yu Chao, Zhili Li, Zhixing Tan, Zhiyuan Liu, Zihan Zhou","submitted_at":"2024-10-12T03:13:44Z","abstract_excerpt":"Enlarging the context window of large language models (LLMs) has become a crucial research area, particularly for applications involving extremely long texts. In this work, we propose a novel training-free framework for processing long texts, utilizing a divide-and-conquer strategy to achieve comprehensive document understanding. The proposed LLM$\\times$MapReduce framework splits the entire document into several chunks for LLMs to read and then aggregates the intermediate answers to produce the final output. The main challenge for divide-and-conquer long text processing frameworks lies in the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.09342","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/2410.09342/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":"2410.09342","created_at":"2026-07-05T09:19:37.941676+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.09342v1","created_at":"2026-07-05T09:19:37.941676+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.09342","created_at":"2026-07-05T09:19:37.941676+00:00"},{"alias_kind":"pith_short_12","alias_value":"2X3FIIGXKDW7","created_at":"2026-07-05T09:19:37.941676+00:00"},{"alias_kind":"pith_short_16","alias_value":"2X3FIIGXKDW7FUQQ","created_at":"2026-07-05T09:19:37.941676+00:00"},{"alias_kind":"pith_short_8","alias_value":"2X3FIIGX","created_at":"2026-07-05T09:19:37.941676+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.23296","citing_title":"Parallel Context Compaction for Long-Horizon LLM Agent Serving","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2510.10129","citing_title":"CacheClip: Accelerating RAG with Effective KV Cache Reuse","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2502.11089","citing_title":"Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU","json":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU.json","graph_json":"https://pith.science/api/pith-number/2X3FIIGXKDW7FUQQHN2EWIWWEU/graph.json","events_json":"https://pith.science/api/pith-number/2X3FIIGXKDW7FUQQHN2EWIWWEU/events.json","paper":"https://pith.science/paper/2X3FIIGX"},"agent_actions":{"view_html":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU","download_json":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU.json","view_paper":"https://pith.science/paper/2X3FIIGX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.09342&json=true","fetch_graph":"https://pith.science/api/pith-number/2X3FIIGXKDW7FUQQHN2EWIWWEU/graph.json","fetch_events":"https://pith.science/api/pith-number/2X3FIIGXKDW7FUQQHN2EWIWWEU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU/action/storage_attestation","attest_author":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU/action/author_attestation","sign_citation":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU/action/citation_signature","submit_replication":"https://pith.science/pith/2X3FIIGXKDW7FUQQHN2EWIWWEU/action/replication_record"}},"created_at":"2026-07-05T09:19:37.941676+00:00","updated_at":"2026-07-05T09:19:37.941676+00:00"}