{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H3KYW26SJ2F3EIQZQ3WUXHW4HD","short_pith_number":"pith:H3KYW26S","schema_version":"1.0","canonical_sha256":"3ed58b6bd24e8bb2221986ed4b9edc38c9455a7e963c0747519df22c99e96d18","source":{"kind":"arxiv","id":"2506.05167","version":2},"attestation_state":"computed","paper":{"title":"ECoRAG: Evidentiality-guided Compression for Long Context RAG","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Dohyeon Lee, Jinsu Kim, Seung-won Hwang, Yeonseok Jeong","submitted_at":"2025-06-05T15:43:49Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable performance in Open-Domain Question Answering (ODQA) by leveraging external documents through Retrieval-Augmented Generation (RAG). To reduce RAG overhead, from longer context, context compression is necessary. However, prior compression methods do not focus on filtering out non-evidential information, which limit the performance in LLM-based RAG. We thus propose Evidentiality-guided RAG, or ECoRAG framework. ECoRAG improves LLM performance by compressing retrieved documents based on evidentiality, ensuring whether answer generation is support"},"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":"2506.05167","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T15:43:49Z","cross_cats_sorted":["cs.AI","cs.IR"],"title_canon_sha256":"fa3d942d30485feb97daeccefd2177aa5a711d42bb50ddf717d2b0c0ef2788da","abstract_canon_sha256":"6ba9584d80ec548984eb1131d8462787816a71a63881fcb0c4e8472e29880b35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:00.856265Z","signature_b64":"Dp2TeORRcz93I9J4jKGmDhkiPN7rzZJBH31znRPtmUjbW7rxNSyjiiJYW2GbdV73SCjHZyCeGpEQtBPRNvs+AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ed58b6bd24e8bb2221986ed4b9edc38c9455a7e963c0747519df22c99e96d18","last_reissued_at":"2026-07-05T11:17:00.855777Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:00.855777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ECoRAG: Evidentiality-guided Compression for Long Context RAG","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.IR"],"primary_cat":"cs.CL","authors_text":"Dohyeon Lee, Jinsu Kim, Seung-won Hwang, Yeonseok Jeong","submitted_at":"2025-06-05T15:43:49Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable performance in Open-Domain Question Answering (ODQA) by leveraging external documents through Retrieval-Augmented Generation (RAG). To reduce RAG overhead, from longer context, context compression is necessary. However, prior compression methods do not focus on filtering out non-evidential information, which limit the performance in LLM-based RAG. We thus propose Evidentiality-guided RAG, or ECoRAG framework. ECoRAG improves LLM performance by compressing retrieved documents based on evidentiality, ensuring whether answer generation is support"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05167","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/2506.05167/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":"2506.05167","created_at":"2026-07-05T11:17:00.855836+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.05167v2","created_at":"2026-07-05T11:17:00.855836+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05167","created_at":"2026-07-05T11:17:00.855836+00:00"},{"alias_kind":"pith_short_12","alias_value":"H3KYW26SJ2F3","created_at":"2026-07-05T11:17:00.855836+00:00"},{"alias_kind":"pith_short_16","alias_value":"H3KYW26SJ2F3EIQZ","created_at":"2026-07-05T11:17:00.855836+00:00"},{"alias_kind":"pith_short_8","alias_value":"H3KYW26S","created_at":"2026-07-05T11:17:00.855836+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/H3KYW26SJ2F3EIQZQ3WUXHW4HD","json":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD.json","graph_json":"https://pith.science/api/pith-number/H3KYW26SJ2F3EIQZQ3WUXHW4HD/graph.json","events_json":"https://pith.science/api/pith-number/H3KYW26SJ2F3EIQZQ3WUXHW4HD/events.json","paper":"https://pith.science/paper/H3KYW26S"},"agent_actions":{"view_html":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD","download_json":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD.json","view_paper":"https://pith.science/paper/H3KYW26S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.05167&json=true","fetch_graph":"https://pith.science/api/pith-number/H3KYW26SJ2F3EIQZQ3WUXHW4HD/graph.json","fetch_events":"https://pith.science/api/pith-number/H3KYW26SJ2F3EIQZQ3WUXHW4HD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD/action/storage_attestation","attest_author":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD/action/author_attestation","sign_citation":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD/action/citation_signature","submit_replication":"https://pith.science/pith/H3KYW26SJ2F3EIQZQ3WUXHW4HD/action/replication_record"}},"created_at":"2026-07-05T11:17:00.855836+00:00","updated_at":"2026-07-05T11:17:00.855836+00:00"}