{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:73Z7U7QJP55DP47NQEUQZTKW4W","short_pith_number":"pith:73Z7U7QJ","schema_version":"1.0","canonical_sha256":"fef3fa7e097f7a37f3ed81290ccd56e58a9dba0e577850041c0471d86e0ad60d","source":{"kind":"arxiv","id":"2306.09539","version":4},"attestation_state":"computed","paper":{"title":"Block-State Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christopher Pal, Jonathan Pilault, Mahan Fathi, Orhan Firat, Pierre-Luc Bacon, Ross Goroshin","submitted_at":"2023-06-15T22:48:08Z","abstract_excerpt":"State space models (SSMs) have shown impressive results on tasks that require modeling long-range dependencies and efficiently scale to long sequences owing to their subquadratic runtime complexity. Originally designed for continuous signals, SSMs have shown superior performance on a plethora of tasks, in vision and audio; however, SSMs still lag Transformer performance in Language Modeling tasks. In this work, we propose a hybrid layer named Block-State Transformer (BST), that internally combines an SSM sublayer for long-range contextualization, and a Block Transformer sublayer for short-term"},"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":"2306.09539","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-15T22:48:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"85a911b3accdd3f0ae39e0ce9952acd9579846fa01eec2a4335b4fa8b618c060","abstract_canon_sha256":"1f7d9b809c3ff3094c645707793d8d51630d465bae452073e310f78d194db7a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:30.265784Z","signature_b64":"NrQ76SvaSPjT8Bw4/fw+3ZodubY9toJOoJSJORGj2dJs9ilf+TOUa47AtSUw3KYDejmcl2TexMaDQEqAJ8B2CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fef3fa7e097f7a37f3ed81290ccd56e58a9dba0e577850041c0471d86e0ad60d","last_reissued_at":"2026-07-05T07:06:30.265293Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:30.265293Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Block-State Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christopher Pal, Jonathan Pilault, Mahan Fathi, Orhan Firat, Pierre-Luc Bacon, Ross Goroshin","submitted_at":"2023-06-15T22:48:08Z","abstract_excerpt":"State space models (SSMs) have shown impressive results on tasks that require modeling long-range dependencies and efficiently scale to long sequences owing to their subquadratic runtime complexity. Originally designed for continuous signals, SSMs have shown superior performance on a plethora of tasks, in vision and audio; however, SSMs still lag Transformer performance in Language Modeling tasks. In this work, we propose a hybrid layer named Block-State Transformer (BST), that internally combines an SSM sublayer for long-range contextualization, and a Block Transformer sublayer for short-term"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09539","kind":"arxiv","version":4},"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/2306.09539/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":"2306.09539","created_at":"2026-07-05T07:06:30.265366+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.09539v4","created_at":"2026-07-05T07:06:30.265366+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09539","created_at":"2026-07-05T07:06:30.265366+00:00"},{"alias_kind":"pith_short_12","alias_value":"73Z7U7QJP55D","created_at":"2026-07-05T07:06:30.265366+00:00"},{"alias_kind":"pith_short_16","alias_value":"73Z7U7QJP55DP47N","created_at":"2026-07-05T07:06:30.265366+00:00"},{"alias_kind":"pith_short_8","alias_value":"73Z7U7QJ","created_at":"2026-07-05T07:06:30.265366+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2312.00752","citing_title":"Mamba: Linear-Time Sequence Modeling with Selective State Spaces","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W","json":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W.json","graph_json":"https://pith.science/api/pith-number/73Z7U7QJP55DP47NQEUQZTKW4W/graph.json","events_json":"https://pith.science/api/pith-number/73Z7U7QJP55DP47NQEUQZTKW4W/events.json","paper":"https://pith.science/paper/73Z7U7QJ"},"agent_actions":{"view_html":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W","download_json":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W.json","view_paper":"https://pith.science/paper/73Z7U7QJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.09539&json=true","fetch_graph":"https://pith.science/api/pith-number/73Z7U7QJP55DP47NQEUQZTKW4W/graph.json","fetch_events":"https://pith.science/api/pith-number/73Z7U7QJP55DP47NQEUQZTKW4W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W/action/storage_attestation","attest_author":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W/action/author_attestation","sign_citation":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W/action/citation_signature","submit_replication":"https://pith.science/pith/73Z7U7QJP55DP47NQEUQZTKW4W/action/replication_record"}},"created_at":"2026-07-05T07:06:30.265366+00:00","updated_at":"2026-07-05T07:06:30.265366+00:00"}