{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:T67M7LZAQDZ5MEVKZWWCNA6MYT","short_pith_number":"pith:T67M7LZA","schema_version":"1.0","canonical_sha256":"9fbecfaf2080f3d612aacdac2683ccc4f606b1459fbc8edd83b4a3a8dd9102d4","source":{"kind":"arxiv","id":"2402.14083","version":2},"attestation_state":"computed","paper":{"title":"Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"DiJia Su, Lucas Lehnert, Michael Rabbat, Paul Mcvay, Qinqing Zheng, Sainbayar Sukhbaatar, Yuandong Tian","submitted_at":"2024-02-21T19:17:28Z","abstract_excerpt":"While Transformers have enabled tremendous progress in various application settings, such architectures still trail behind traditional symbolic planners for solving complex decision making tasks. In this work, we demonstrate how to train Transformers to solve complex planning tasks. This is accomplished by training an encoder-decoder Transformer model to predict the search dynamics of the $A^*$ search algorithm. We fine tune this model to obtain a Searchformer, a Transformer model that optimally solves previously unseen Sokoban puzzles 93.7% of the time, while using up to 26.8% fewer search st"},"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":"2402.14083","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-02-21T19:17:28Z","cross_cats_sorted":[],"title_canon_sha256":"7c1e871a3074115ba50dd4a4170c8bd9bbc701e9436e1b1230acdc3d4b03d86f","abstract_canon_sha256":"a2a51befa063f684ae4cd6d8358f4060f4be6c695adafa0b92bfca70559ada73"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:12:48.421374Z","signature_b64":"KDDMMYOJt/pBk2QFEaGHcifNtIatces7rQRDG44zk9QT+H2OenkVc9vN/PutlpoLIvRy4AlxYG5nTroe3LqbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fbecfaf2080f3d612aacdac2683ccc4f606b1459fbc8edd83b4a3a8dd9102d4","last_reissued_at":"2026-07-05T08:12:48.420959Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:12:48.420959Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"DiJia Su, Lucas Lehnert, Michael Rabbat, Paul Mcvay, Qinqing Zheng, Sainbayar Sukhbaatar, Yuandong Tian","submitted_at":"2024-02-21T19:17:28Z","abstract_excerpt":"While Transformers have enabled tremendous progress in various application settings, such architectures still trail behind traditional symbolic planners for solving complex decision making tasks. In this work, we demonstrate how to train Transformers to solve complex planning tasks. This is accomplished by training an encoder-decoder Transformer model to predict the search dynamics of the $A^*$ search algorithm. We fine tune this model to obtain a Searchformer, a Transformer model that optimally solves previously unseen Sokoban puzzles 93.7% of the time, while using up to 26.8% fewer search st"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14083","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/2402.14083/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":"2402.14083","created_at":"2026-07-05T08:12:48.421018+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.14083v2","created_at":"2026-07-05T08:12:48.421018+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14083","created_at":"2026-07-05T08:12:48.421018+00:00"},{"alias_kind":"pith_short_12","alias_value":"T67M7LZAQDZ5","created_at":"2026-07-05T08:12:48.421018+00:00"},{"alias_kind":"pith_short_16","alias_value":"T67M7LZAQDZ5MEVK","created_at":"2026-07-05T08:12:48.421018+00:00"},{"alias_kind":"pith_short_8","alias_value":"T67M7LZA","created_at":"2026-07-05T08:12:48.421018+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07492","citing_title":"Search, Fail, Recover: A Training Framework for Correction-Aware Reasoning","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00618","citing_title":"Efficient Test-time Inference for Generative Planning Models with OCL Search","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06556","citing_title":"Robots Need More than VLA and World Models","ref_index":149,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18820","citing_title":"Emergence of Frontier Superposition: M\\\"obius attractor and Cascade Supervision","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2408.07199","citing_title":"Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04476","citing_title":"Vision-aligned Latent Reasoning for Multi-modal Large Language Model","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2510.04871","citing_title":"Less is More: Recursive Reasoning with Tiny Networks","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03625","citing_title":"Self-Improvement for Fast, High-Quality Plan Generation","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08221","citing_title":"NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13151","citing_title":"Exploration and Exploitation Errors Are Measurable for Language Model Agents","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06769","citing_title":"Training Large Language Models to Reason in a Continuous Latent Space","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT","json":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT.json","graph_json":"https://pith.science/api/pith-number/T67M7LZAQDZ5MEVKZWWCNA6MYT/graph.json","events_json":"https://pith.science/api/pith-number/T67M7LZAQDZ5MEVKZWWCNA6MYT/events.json","paper":"https://pith.science/paper/T67M7LZA"},"agent_actions":{"view_html":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT","download_json":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT.json","view_paper":"https://pith.science/paper/T67M7LZA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.14083&json=true","fetch_graph":"https://pith.science/api/pith-number/T67M7LZAQDZ5MEVKZWWCNA6MYT/graph.json","fetch_events":"https://pith.science/api/pith-number/T67M7LZAQDZ5MEVKZWWCNA6MYT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT/action/storage_attestation","attest_author":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT/action/author_attestation","sign_citation":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT/action/citation_signature","submit_replication":"https://pith.science/pith/T67M7LZAQDZ5MEVKZWWCNA6MYT/action/replication_record"}},"created_at":"2026-07-05T08:12:48.421018+00:00","updated_at":"2026-07-05T08:12:48.421018+00:00"}