{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GS5JCQOGJWBZWODM6RWRVP5CDI","short_pith_number":"pith:GS5JCQOG","schema_version":"1.0","canonical_sha256":"34ba9141c64d839b386cf46d1abfa21a1fc7fcf42c8924189e49917727cae8dd","source":{"kind":"arxiv","id":"2402.16617","version":2},"attestation_state":"computed","paper":{"title":"Long-Context Language Modeling with Parallel Context Encoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danqi Chen, Howard Yen, Tianyu Gao","submitted_at":"2024-02-26T14:47:35Z","abstract_excerpt":"Extending large language models (LLMs) to process longer inputs is crucial for a wide range of applications. However, the substantial computational cost of transformers and limited generalization of positional encoding restrict the size of their context window. We introduce Context Expansion with Parallel Encoding (CEPE), a framework that can be applied to any existing decoder-only LLMs to extend their context window. CEPE employs a small encoder to process long inputs chunk by chunk, enabling the frozen decoder to utilize additional contexts via cross-attention. CEPE is efficient, generalizab"},"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":"2402.16617","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-26T14:47:35Z","cross_cats_sorted":[],"title_canon_sha256":"c6c405e5656e25f42b2ef311ba1a826f2d4fa362244638daa05f8d4c1a6baf7e","abstract_canon_sha256":"5ab4de74096b0fd6170424534a25daf0cd07e5bddb27e8a2f89fd669cfb07a32"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:57.818742Z","signature_b64":"OIZVZzrdR/TiqE0HKLnOvupXyhpsaZXoIKcIpgLpEWG+8GF8xZfWdEquBVRtHk1JkbByXrIbld7BWp0ZUc4tCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34ba9141c64d839b386cf46d1abfa21a1fc7fcf42c8924189e49917727cae8dd","last_reissued_at":"2026-07-05T11:18:57.818253Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:57.818253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Long-Context Language Modeling with Parallel Context Encoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danqi Chen, Howard Yen, Tianyu Gao","submitted_at":"2024-02-26T14:47:35Z","abstract_excerpt":"Extending large language models (LLMs) to process longer inputs is crucial for a wide range of applications. However, the substantial computational cost of transformers and limited generalization of positional encoding restrict the size of their context window. We introduce Context Expansion with Parallel Encoding (CEPE), a framework that can be applied to any existing decoder-only LLMs to extend their context window. CEPE employs a small encoder to process long inputs chunk by chunk, enabling the frozen decoder to utilize additional contexts via cross-attention. CEPE is efficient, generalizab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.16617","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/2402.16617/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":"2402.16617","created_at":"2026-07-05T11:18:57.818322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.16617v2","created_at":"2026-07-05T11:18:57.818322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.16617","created_at":"2026-07-05T11:18:57.818322+00:00"},{"alias_kind":"pith_short_12","alias_value":"GS5JCQOGJWBZ","created_at":"2026-07-05T11:18:57.818322+00:00"},{"alias_kind":"pith_short_16","alias_value":"GS5JCQOGJWBZWODM","created_at":"2026-07-05T11:18:57.818322+00:00"},{"alias_kind":"pith_short_8","alias_value":"GS5JCQOG","created_at":"2026-07-05T11:18:57.818322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17041","citing_title":"MetaSyn: A Benchmark for LLM Agents on Meta-Analysis Articles from Nature Portfolio","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2510.22102","citing_title":"Mitigating Coordinate Prediction Bias from Positional Encoding Failures","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04595","citing_title":"A Queueing-Theoretic Framework for Stability Analysis of LLM Inference with KV Cache Memory Constraints","ref_index":117,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI","json":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI.json","graph_json":"https://pith.science/api/pith-number/GS5JCQOGJWBZWODM6RWRVP5CDI/graph.json","events_json":"https://pith.science/api/pith-number/GS5JCQOGJWBZWODM6RWRVP5CDI/events.json","paper":"https://pith.science/paper/GS5JCQOG"},"agent_actions":{"view_html":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI","download_json":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI.json","view_paper":"https://pith.science/paper/GS5JCQOG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.16617&json=true","fetch_graph":"https://pith.science/api/pith-number/GS5JCQOGJWBZWODM6RWRVP5CDI/graph.json","fetch_events":"https://pith.science/api/pith-number/GS5JCQOGJWBZWODM6RWRVP5CDI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI/action/storage_attestation","attest_author":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI/action/author_attestation","sign_citation":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI/action/citation_signature","submit_replication":"https://pith.science/pith/GS5JCQOGJWBZWODM6RWRVP5CDI/action/replication_record"}},"created_at":"2026-07-05T11:18:57.818322+00:00","updated_at":"2026-07-05T11:18:57.818322+00:00"}