{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QY5J6ZEFJMY7G5JISBITUL63VB","short_pith_number":"pith:QY5J6ZEF","schema_version":"1.0","canonical_sha256":"863a9f64854b31f3752890513a2fdba87e55adfe4e28126a59d30305b32a5732","source":{"kind":"arxiv","id":"2606.10621","version":1},"attestation_state":"computed","paper":{"title":"STORM: Stepwise Token Optimization with Reward-Guided Beam Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Arthur Satouf, Benjamin Piwowarski, Giulio D'Erasmo, Habiboulaye Amadou Boubacar, Pablo Piantanida, Yuxuan Zong","submitted_at":"2026-06-09T09:18:09Z","abstract_excerpt":"Modern retrieval increasingly relies on dense and learned-sparse neural models that are effective but require encoding the entire corpus into a specialized index, rebuilt whenever the model changes. Lexical retrievers like BM25 stay efficient and transparent on a standard inverted index that need not change as models evolve, but suffer from vocabulary mismatch. LLM query rewriting can help, yet prompted rewriters emit well-formed but retrieval-ineffective or harmful-terms, and training against a retrieval reward gives only delayed, sequence-level supervision that obscures which terms helped. W"},"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":"2606.10621","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-06-09T09:18:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"53cdbcb2fd92fbd585d4fc3aa26f62829807f27e6b7adf910476b7d012d032c8","abstract_canon_sha256":"01d8b2f5b2df8d78f324ca5916c6214126831a38e71fa63ef3f566bd5f97e4a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:10:30.100207Z","signature_b64":"ZbzBE41A/j0Z7be65CCL3lm4E4crL3sZ19j8x3Wr4gUVK05FLP8eEqBnWdApjhXVEO4k7CgH48SalY7bO8gpCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"863a9f64854b31f3752890513a2fdba87e55adfe4e28126a59d30305b32a5732","last_reissued_at":"2026-06-10T01:10:30.099337Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:10:30.099337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"STORM: Stepwise Token Optimization with Reward-Guided Beam Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.IR","authors_text":"Arthur Satouf, Benjamin Piwowarski, Giulio D'Erasmo, Habiboulaye Amadou Boubacar, Pablo Piantanida, Yuxuan Zong","submitted_at":"2026-06-09T09:18:09Z","abstract_excerpt":"Modern retrieval increasingly relies on dense and learned-sparse neural models that are effective but require encoding the entire corpus into a specialized index, rebuilt whenever the model changes. Lexical retrievers like BM25 stay efficient and transparent on a standard inverted index that need not change as models evolve, but suffer from vocabulary mismatch. LLM query rewriting can help, yet prompted rewriters emit well-formed but retrieval-ineffective or harmful-terms, and training against a retrieval reward gives only delayed, sequence-level supervision that obscures which terms helped. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10621","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/2606.10621/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":"2606.10621","created_at":"2026-06-10T01:10:30.099478+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.10621v1","created_at":"2026-06-10T01:10:30.099478+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10621","created_at":"2026-06-10T01:10:30.099478+00:00"},{"alias_kind":"pith_short_12","alias_value":"QY5J6ZEFJMY7","created_at":"2026-06-10T01:10:30.099478+00:00"},{"alias_kind":"pith_short_16","alias_value":"QY5J6ZEFJMY7G5JI","created_at":"2026-06-10T01:10:30.099478+00:00"},{"alias_kind":"pith_short_8","alias_value":"QY5J6ZEF","created_at":"2026-06-10T01:10:30.099478+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/QY5J6ZEFJMY7G5JISBITUL63VB","json":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB.json","graph_json":"https://pith.science/api/pith-number/QY5J6ZEFJMY7G5JISBITUL63VB/graph.json","events_json":"https://pith.science/api/pith-number/QY5J6ZEFJMY7G5JISBITUL63VB/events.json","paper":"https://pith.science/paper/QY5J6ZEF"},"agent_actions":{"view_html":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB","download_json":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB.json","view_paper":"https://pith.science/paper/QY5J6ZEF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.10621&json=true","fetch_graph":"https://pith.science/api/pith-number/QY5J6ZEFJMY7G5JISBITUL63VB/graph.json","fetch_events":"https://pith.science/api/pith-number/QY5J6ZEFJMY7G5JISBITUL63VB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB/action/storage_attestation","attest_author":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB/action/author_attestation","sign_citation":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB/action/citation_signature","submit_replication":"https://pith.science/pith/QY5J6ZEFJMY7G5JISBITUL63VB/action/replication_record"}},"created_at":"2026-06-10T01:10:30.099478+00:00","updated_at":"2026-06-10T01:10:30.099478+00:00"}