{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:762NPQJWUI74ONW4WKKRJ5HFWK","short_pith_number":"pith:762NPQJW","schema_version":"1.0","canonical_sha256":"ffb4d7c136a23fc736dcb29514f4e5b2a00a8b2c56f1b0a34b3ca23515404b14","source":{"kind":"arxiv","id":"2511.05385","version":2},"attestation_state":"computed","paper":{"title":"TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Chao Zhang, Derong Xu, Enhong Chen, Haoxin Zhang, Shuochen Liu, Tong Xu, Xiangyu Zhao, Yan Gao, Yao Hu, Yuanjie Lyu, Yuhao Chen, Yuhao Wang","submitted_at":"2025-11-07T16:08:34Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) utilizes external knowledge to augment Large Language Models' (LLMs) reliability. For flexibility, agentic RAG employs autonomous, multi-round retrieval and reasoning to resolve queries. Although recent agentic RAG has improved via reinforcement learning, they often incur substantial token overhead from search and reasoning processes. This trade-off prioritizes accuracy over efficiency. To address this issue, this work proposes TeaRAG, a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps. 1) First, the re"},"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":"2511.05385","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2025-11-07T16:08:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"785c6a6c023877fefcdd25be673d8141f84fb5a3b9dd190c316e21fc44383bd7","abstract_canon_sha256":"11d5cd80f024b80116e85a6f44059abdad84fbabf3c71081ff5cba62a7be5eb4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:23:01.229717Z","signature_b64":"1ahQVm/HeeaBrmQB57oDTVZDUGYMW5TcPR+LBJ0SiZ6QUidlXQ1j6djNu1vOGvDLJ8Gs1kkuYg+cf6K7PgJpDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ffb4d7c136a23fc736dcb29514f4e5b2a00a8b2c56f1b0a34b3ca23515404b14","last_reissued_at":"2026-07-24T01:23:01.228768Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:23:01.228768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Chao Zhang, Derong Xu, Enhong Chen, Haoxin Zhang, Shuochen Liu, Tong Xu, Xiangyu Zhao, Yan Gao, Yao Hu, Yuanjie Lyu, Yuhao Chen, Yuhao Wang","submitted_at":"2025-11-07T16:08:34Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) utilizes external knowledge to augment Large Language Models' (LLMs) reliability. For flexibility, agentic RAG employs autonomous, multi-round retrieval and reasoning to resolve queries. Although recent agentic RAG has improved via reinforcement learning, they often incur substantial token overhead from search and reasoning processes. This trade-off prioritizes accuracy over efficiency. To address this issue, this work proposes TeaRAG, a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps. 1) First, the re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.05385","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/2511.05385/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":"2511.05385","created_at":"2026-07-24T01:23:01.229209+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.05385v2","created_at":"2026-07-24T01:23:01.229209+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.05385","created_at":"2026-07-24T01:23:01.229209+00:00"},{"alias_kind":"pith_short_12","alias_value":"762NPQJWUI74","created_at":"2026-07-24T01:23:01.229209+00:00"},{"alias_kind":"pith_short_16","alias_value":"762NPQJWUI74ONW4","created_at":"2026-07-24T01:23:01.229209+00:00"},{"alias_kind":"pith_short_8","alias_value":"762NPQJW","created_at":"2026-07-24T01:23:01.229209+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2603.23231","citing_title":"PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05538","citing_title":"AgenticRAG: Agentic Retrieval for Enterprise Knowledge Bases","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK","json":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK.json","graph_json":"https://pith.science/api/pith-number/762NPQJWUI74ONW4WKKRJ5HFWK/graph.json","events_json":"https://pith.science/api/pith-number/762NPQJWUI74ONW4WKKRJ5HFWK/events.json","paper":"https://pith.science/paper/762NPQJW"},"agent_actions":{"view_html":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK","download_json":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK.json","view_paper":"https://pith.science/paper/762NPQJW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.05385&json=true","fetch_graph":"https://pith.science/api/pith-number/762NPQJWUI74ONW4WKKRJ5HFWK/graph.json","fetch_events":"https://pith.science/api/pith-number/762NPQJWUI74ONW4WKKRJ5HFWK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK/action/storage_attestation","attest_author":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK/action/author_attestation","sign_citation":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK/action/citation_signature","submit_replication":"https://pith.science/pith/762NPQJWUI74ONW4WKKRJ5HFWK/action/replication_record"}},"created_at":"2026-07-24T01:23:01.229209+00:00","updated_at":"2026-07-24T01:23:01.229209+00:00"}