{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","short_pith_number":"pith:KNYZYJAU","canonical_record":{"source":{"id":"2605.06647","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-07T17:54:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"add8bfec8a133273e06abc9dd5b1ff97cdfffe33da8988eb36d2eed083c6cd6f","abstract_canon_sha256":"3168eaab97a43939f979ec5c93e54dbff134d8dc33ddbd943564f53f16117095"},"schema_version":"1.0"},"canonical_sha256":"53719c24145f67cfbe7eeaab3663e6edcb69db0f615a46571f7768d1d0cca88b","source":{"kind":"arxiv","id":"2605.06647","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06647","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06647v2","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06647","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_12","alias_value":"KNYZYJAUL5T4","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_16","alias_value":"KNYZYJAUL5T47PT6","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_8","alias_value":"KNYZYJAU","created_at":"2026-06-08T01:04:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","target":"record","payload":{"canonical_record":{"source":{"id":"2605.06647","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-07T17:54:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"add8bfec8a133273e06abc9dd5b1ff97cdfffe33da8988eb36d2eed083c6cd6f","abstract_canon_sha256":"3168eaab97a43939f979ec5c93e54dbff134d8dc33ddbd943564f53f16117095"},"schema_version":"1.0"},"canonical_sha256":"53719c24145f67cfbe7eeaab3663e6edcb69db0f615a46571f7768d1d0cca88b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-08T01:04:06.785170Z","signature_b64":"AooEcsupEt2SFEALLJ99bmxWi5wXWk2G5WeeaGWhWbhGA7S7np/GU0ocL8CigTaJDaNW0AALditImj7lIxfBDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53719c24145f67cfbe7eeaab3663e6edcb69db0f615a46571f7768d1d0cca88b","last_reissued_at":"2026-06-08T01:04:06.784312Z","signature_status":"signed_v1","first_computed_at":"2026-06-08T01:04:06.784312Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.06647","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-08T01:04:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nyRr//UdR9izlPxXD1FSGpRCgPYe+9Pu47h5if5lilN2YqR47PWsujrlAM9a9Cjwvqnsw1eccWr3kw2ElFfSAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T21:51:02.635634Z"},"content_sha256":"70b2e0cd008d659e476419fed353ffe5f08ed95a7c6ce76a41bb54f858b0dccd","schema_version":"1.0","event_id":"sha256:70b2e0cd008d659e476419fed353ffe5f08ed95a7c6ce76a41bb54f858b0dccd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Superintelligent Retrieval Agent: The Next Frontier of Agentic Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.IR","authors_text":"Anshumali Shrivastava, Jason Chen, Qi Ma, Zeyu Yang","submitted_at":"2026-05-07T17:54:29Z","abstract_excerpt":"Retrieval-augmented agents are increasingly the interface to large knowledge bases, yet most treat retrieval as a black box: they issue exploratory queries, inspect snippets, and reformulate until evidence emerges. This resembles how a newcomer searches an unfamiliar database rather than how an expert navigates it with strong priors about terminology and likely evidence, causing extra retrieval rounds, latency, and poor recall.\n  We introduce \\textit{Superintelligent Retrieval Agent} (SIRA), which casts \\emph{superintelligence} in retrieval as compressing multi-round exploratory search into a "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across ten BEIR benchmarks and downstream question-answering tasks, SIRA achieves the significantly superior performance outperforming dense retrievers and state-of-the-art multi-round agentic baselines, demonstrating that one well-formed lexical query, guided by LLM cognition and lightweight corpus statistics, can exceed substantially more expensive multi-round search.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That LLM-generated term predictions and document-frequency filtering will reliably produce terms that increase retrieval margin without adding noise or omitting critical evidence, and that this advantage holds across diverse corpora and query types.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SIRA compresses multi-round exploratory retrieval into one LLM-guided, corpus-statistic-validated weighted BM25 query and reports superior results over dense retrievers and agentic baselines on BEIR benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"b7835958f355457fe325cd51ba201bd2767398c03b01692f68449dd7adf177fb"},"source":{"id":"2605.06647","kind":"arxiv","version":2},"verdict":{"id":"f696af1f-8126-4d3e-a3c3-3bf279fc945c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T05:36:43.716955Z","strongest_claim":"Across ten BEIR benchmarks and downstream question-answering tasks, SIRA achieves the significantly superior performance outperforming dense retrievers and state-of-the-art multi-round agentic baselines, demonstrating that one well-formed lexical query, guided by LLM cognition and lightweight corpus statistics, can exceed substantially more expensive multi-round search.","one_line_summary":"SIRA compresses multi-round exploratory retrieval into one LLM-guided, corpus-statistic-validated weighted BM25 query and reports superior results over dense retrievers and agentic baselines on BEIR benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That LLM-generated term predictions and document-frequency filtering will reliably produce terms that increase retrieval margin without adding noise or omitting critical evidence, and that this advantage holds across diverse corpora and query types.","pith_extraction_headline":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers."},"integrity":{"clean":false,"summary":{"advisory":2,"critical":0,"by_detector":{"doi_compliance":{"total":2,"advisory":2,"critical":0,"informational":0}},"informational":0},"endpoint":"/pith/2605.06647/integrity.json","findings":[{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":4,"audited_at":"2026-05-19T12:31:59.292711Z","detected_doi":"10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null},{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":11,"audited_at":"2026-05-19T12:31:59.292711Z","detected_doi":"10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T12:02:03.897858Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T07:37:07.346750Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T18:01:19.201073Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:31:59.292711Z","status":"completed","version":"1.0.0","findings_count":2}],"snapshot_sha256":"ff9a95acf358eb90c3b754ba9367f48b6a8664aa1a36f4f8de59e8ae27eddffa"},"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"},"verdict_id":"f696af1f-8126-4d3e-a3c3-3bf279fc945c"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-08T01:04:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/E8jp+Ctb6RU6QxJvwGB5/WhSmYgF6NIMGkLc5GCOFmjTk/AR36ZonxzmN6CUYjmquUBPUBy1g0IrldWhmIBCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T21:51:02.636454Z"},"content_sha256":"e58f4e82aca21e72b11bc3daed94720c12ea0534b2f16c91421288654e4d575c","schema_version":"1.0","event_id":"sha256:e58f4e82aca21e72b11bc3daed94720c12ea0534b2f16c91421288654e4d575c"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Association for Computational Linguistics. doi: 10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/. Baoyi Wang, Xingliang Wang, Guochang Li, Chen Zhi, Junxiao Han, Xinkui Zhao, Nan Wang, Shuiguang Deng, and Jianwei Yi","arxiv_id":"2605.06647","detector":"doi_compliance","evidence":{"ref_index":11,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Association for Computational Linguistics. doi: 10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/. Baoyi Wang, Xingliang Wang, Guochang Li, Chen Zhi, Junxiao Han, Xinkui Zhao, Nan Wang, Shuiguang Deng, and Jianwei Yi","reconstructed_doi":"10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi"},"severity":"advisory","ref_index":11,"audited_at":"2026-05-19T12:31:59.292711Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"216be21efa0a9333521ddd0adfc6a60a3d03cf7a796bdaa396793100d0ccf40a","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":1201,"payload_sha256":"763a99375057ad07164eb0be697e2dd79f0e6f8ec66730d941202156a645f1c5","signature_b64":"bRG9u3rVZrD9tRIGpcGp6A/r7AGEM55pshLeb375el54KsOc4zbl1hZZ4Yg7sUdbeTOcWBrQp3rQZOMeCSYUAQ==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T12:32:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RpntARG/qrswk3UB/XbrAUT9/LkM1BnV1GqcNHjVVHl2kfX5HG492nK5D/R2guFO2OzG0kjG1RTpk+AVdZOmAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T21:51:02.640726Z"},"content_sha256":"f3e8a84b8aedef05fc3d2fca4becb509d05e742cac09ecc8021e3b04bf178c57","schema_version":"1.0","event_id":"sha256:f3e8a84b8aedef05fc3d2fca4becb509d05e742cac09ecc8021e3b04bf178c57"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"doi: 10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/. Hongliang Lu, Yuhang Wen, Pengyu Cheng, Ruijin Ding, Jiaqi Guo, Haotian Xu, Chutian Wang, Haonan Chen, Xiaoxi Jiang, and Guanjun Jiang. Search self-play: Pushing the frontie","arxiv_id":"2605.06647","detector":"doi_compliance","evidence":{"ref_index":4,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"doi: 10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/. Hongliang Lu, Yuhang Wen, Pengyu Cheng, Ruijin Ding, Jiaqi Guo, Haotian Xu, Chutian Wang, Haonan Chen, Xiaoxi Jiang, and Guanjun Jiang. Search self-play: Pushing the frontie","reconstructed_doi":"10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang"},"severity":"advisory","ref_index":4,"audited_at":"2026-05-19T12:31:59.292711Z","event_type":"pith.integrity.v1","detected_doi":"10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"01cef6bf02db0d18811c1979fad65607753856eaf79f78aed8a86c702c4c173f","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":1200,"payload_sha256":"ea136258a3aabb5ef7109a98fab7613852a0b254de7d06b332c888fcfd62e063","signature_b64":"7sJfkHB4pKJMO9NzSMX70ZpKN5OLdgo+CeSxm5RONlm2+R5a9Oc7UBCpbl2l7as6BXBEuP59mQCdq8OBX+96Cg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T12:32:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AYLOT3quutTzVcrMcVYIoHlSEdZcPrjEZ0JUtPmvE6uO8EA30W4TjMdhhu61VWxREe/l1pAOUv11hc/8fzKSCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T21:51:02.641198Z"},"content_sha256":"0149c1e1d62c8acff4bc3d202b2ecf273882996cce240ad012cbbb1e061ba353","schema_version":"1.0","event_id":"sha256:0149c1e1d62c8acff4bc3d202b2ecf273882996cce240ad012cbbb1e061ba353"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KNYZYJAUL5T47PT65KVTMY7G5X/bundle.json","state_url":"https://pith.science/pith/KNYZYJAUL5T47PT65KVTMY7G5X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KNYZYJAUL5T47PT65KVTMY7G5X/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-21T21:51:02Z","links":{"resolver":"https://pith.science/pith/KNYZYJAUL5T47PT65KVTMY7G5X","bundle":"https://pith.science/pith/KNYZYJAUL5T47PT65KVTMY7G5X/bundle.json","state":"https://pith.science/pith/KNYZYJAUL5T47PT65KVTMY7G5X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KNYZYJAUL5T47PT65KVTMY7G5X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KNYZYJAUL5T47PT65KVTMY7G5X","merge_version":"pith-open-graph-merge-v1","event_count":4,"valid_event_count":4,"invalid_event_count":0,"equivocation_count":1,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3168eaab97a43939f979ec5c93e54dbff134d8dc33ddbd943564f53f16117095","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-07T17:54:29Z","title_canon_sha256":"add8bfec8a133273e06abc9dd5b1ff97cdfffe33da8988eb36d2eed083c6cd6f"},"schema_version":"1.0","source":{"id":"2605.06647","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06647","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06647v2","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06647","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_12","alias_value":"KNYZYJAUL5T4","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_16","alias_value":"KNYZYJAUL5T47PT6","created_at":"2026-06-08T01:04:06Z"},{"alias_kind":"pith_short_8","alias_value":"KNYZYJAU","created_at":"2026-06-08T01:04:06Z"}],"graph_snapshots":[{"event_id":"sha256:e58f4e82aca21e72b11bc3daed94720c12ea0534b2f16c91421288654e4d575c","target":"graph","created_at":"2026-06-08T01:04:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Across ten BEIR benchmarks and downstream question-answering tasks, SIRA achieves the significantly superior performance outperforming dense retrievers and state-of-the-art multi-round agentic baselines, demonstrating that one well-formed lexical query, guided by LLM cognition and lightweight corpus statistics, can exceed substantially more expensive multi-round search."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That LLM-generated term predictions and document-frequency filtering will reliably produce terms that increase retrieval margin without adding noise or omitting critical evidence, and that this advantage holds across diverse corpora and query types."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SIRA compresses multi-round exploratory retrieval into one LLM-guided, corpus-statistic-validated weighted BM25 query and reports superior results over dense retrievers and agentic baselines on BEIR benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers."}],"snapshot_sha256":"b7835958f355457fe325cd51ba201bd2767398c03b01692f68449dd7adf177fb"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":false,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T12:02:03.897858Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T07:37:07.346750Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T18:01:19.201073Z","status":"completed","version":"1.0.0"},{"findings_count":2,"name":"doi_compliance","ran_at":"2026-05-19T12:31:59.292711Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.06647/integrity.json","findings":[{"audited_at":"2026-05-19T12:31:59.292711Z","detected_arxiv_id":null,"detected_doi":"10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang","detector":"doi_compliance","finding_type":"recoverable_identifier","note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1162/tacl_a_00638.https://aclanthology.org/2024.tacl-1.9/.Hongliang) was visible in the surrounding text but could not be confirmed against doi.org as printed.","ref_index":4,"severity":"advisory","verdict_class":"incontrovertible"},{"audited_at":"2026-05-19T12:31:59.292711Z","detected_arxiv_id":null,"detected_doi":"10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi","detector":"doi_compliance","finding_type":"recoverable_identifier","note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2023.acl-long.557.https://aclanthology.org/2023.acl-long.557/.Baoyi) was visible in the surrounding text but could not be confirmed against doi.org as printed.","ref_index":11,"severity":"advisory","verdict_class":"incontrovertible"}],"snapshot_sha256":"ff9a95acf358eb90c3b754ba9367f48b6a8664aa1a36f4f8de59e8ae27eddffa","summary":{"advisory":2,"by_detector":{"doi_compliance":{"advisory":2,"critical":0,"informational":0,"total":2}},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Retrieval-augmented agents are increasingly the interface to large knowledge bases, yet most treat retrieval as a black box: they issue exploratory queries, inspect snippets, and reformulate until evidence emerges. This resembles how a newcomer searches an unfamiliar database rather than how an expert navigates it with strong priors about terminology and likely evidence, causing extra retrieval rounds, latency, and poor recall.\n  We introduce \\textit{Superintelligent Retrieval Agent} (SIRA), which casts \\emph{superintelligence} in retrieval as compressing multi-round exploratory search into a ","authors_text":"Anshumali Shrivastava, Jason Chen, Qi Ma, Zeyu Yang","cross_cats":["cs.AI","cs.LG"],"headline":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-07T17:54:29Z","title":"Superintelligent Retrieval Agent: The Next Frontier of Agentic Retrieval"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.06647","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T05:36:43.716955Z","id":"f696af1f-8126-4d3e-a3c3-3bf279fc945c","model_set":{"reader":"grok-4.3"},"one_line_summary":"SIRA compresses multi-round exploratory retrieval into one LLM-guided, corpus-statistic-validated weighted BM25 query and reports superior results over dense retrievers and agentic baselines on BEIR benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A single LLM-guided lexical query with corpus statistics can outperform multi-round retrieval agents and dense retrievers.","strongest_claim":"Across ten BEIR benchmarks and downstream question-answering tasks, SIRA achieves the significantly superior performance outperforming dense retrievers and state-of-the-art multi-round agentic baselines, demonstrating that one well-formed lexical query, guided by LLM cognition and lightweight corpus statistics, can exceed substantially more expensive multi-round search.","weakest_assumption":"That LLM-generated term predictions and document-frequency filtering will reliably produce terms that increase retrieval margin without adding noise or omitting critical evidence, and that this advantage holds across diverse corpora and query types."}},"verdict_id":"f696af1f-8126-4d3e-a3c3-3bf279fc945c"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:70b2e0cd008d659e476419fed353ffe5f08ed95a7c6ce76a41bb54f858b0dccd","target":"record","created_at":"2026-06-08T01:04:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3168eaab97a43939f979ec5c93e54dbff134d8dc33ddbd943564f53f16117095","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-05-07T17:54:29Z","title_canon_sha256":"add8bfec8a133273e06abc9dd5b1ff97cdfffe33da8988eb36d2eed083c6cd6f"},"schema_version":"1.0","source":{"id":"2605.06647","kind":"arxiv","version":2}},"canonical_sha256":"53719c24145f67cfbe7eeaab3663e6edcb69db0f615a46571f7768d1d0cca88b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"53719c24145f67cfbe7eeaab3663e6edcb69db0f615a46571f7768d1d0cca88b","first_computed_at":"2026-06-08T01:04:06.784312Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-08T01:04:06.784312Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AooEcsupEt2SFEALLJ99bmxWi5wXWk2G5WeeaGWhWbhGA7S7np/GU0ocL8CigTaJDaNW0AALditImj7lIxfBDg==","signature_status":"signed_v1","signed_at":"2026-06-08T01:04:06.785170Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.06647","source_kind":"arxiv","source_version":2}}},"equivocations":[{"signer_id":"pith.science","event_type":"integrity_finding","target":"integrity","event_ids":["sha256:0149c1e1d62c8acff4bc3d202b2ecf273882996cce240ad012cbbb1e061ba353","sha256:f3e8a84b8aedef05fc3d2fca4becb509d05e742cac09ecc8021e3b04bf178c57"]}],"invalid_events":[],"applied_event_ids":["sha256:70b2e0cd008d659e476419fed353ffe5f08ed95a7c6ce76a41bb54f858b0dccd","sha256:e58f4e82aca21e72b11bc3daed94720c12ea0534b2f16c91421288654e4d575c"],"state_sha256":"022399cc2b9520dac53aa53e39cb93e32e4aecedb54a16f0602f8192140435c8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m1QoJ3/WYkZi6BrnlcMm6BprkVEYFOkQnYfyzvpB7givC4EIe7ignq0LieT79lVWFKFtHZX5mjLBRufc2+wnAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T21:51:02.644492Z","bundle_sha256":"be51649bd55e43e2699ae2448853ec4d2734c45aee91bfbad14c1e9a34cd5d9d"}}