{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TTUABZN2SB4C5J2KCXUMQNEEGU","short_pith_number":"pith:TTUABZN2","schema_version":"1.0","canonical_sha256":"9ce800e5ba90782ea74a15e8c834843537f86a587cd8c973f802a7a6a057c72c","source":{"kind":"arxiv","id":"2404.03683","version":1},"attestation_state":"computed","paper":{"title":"Stream of Search (SoS): Learning to Search in Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Archit Sharma, Denise Lee, Gabriel Grand, Kanishk Gandhi, Muxin Liu, Noah D. Goodman, Winson Cheng","submitted_at":"2024-04-01T06:50:52Z","abstract_excerpt":"Language models are rarely shown fruitful mistakes while training. They then struggle to look beyond the next token, suffering from a snowballing of errors and struggling to predict the consequence of their actions several steps ahead. In this paper, we show how language models can be taught to search by representing the process of search in language, as a flattened string -- a stream of search (SoS). We propose a unified language for search that captures an array of different symbolic search strategies. We demonstrate our approach using the simple yet difficult game of Countdown, where the go"},"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":"2404.03683","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-01T06:50:52Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"f1b01ee86e7cd8e42696fcce77570f743d3f721b33fa0373de759d7e8ac06e31","abstract_canon_sha256":"53b8e8e9a60bc9233d784b44ea39f32204d657caa77828d126e849f1aefcbb57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:04:39.968423Z","signature_b64":"TRZ/i730qn52h4hGlAqxrYywb4zC66bwCErJ02qXJ+FQhIcHqC8qclLK8ScIZH7Ghftuo1im4n5EtvLLKeP9CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ce800e5ba90782ea74a15e8c834843537f86a587cd8c973f802a7a6a057c72c","last_reissued_at":"2026-07-05T08:04:39.967860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:04:39.967860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stream of Search (SoS): Learning to Search in Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Archit Sharma, Denise Lee, Gabriel Grand, Kanishk Gandhi, Muxin Liu, Noah D. Goodman, Winson Cheng","submitted_at":"2024-04-01T06:50:52Z","abstract_excerpt":"Language models are rarely shown fruitful mistakes while training. They then struggle to look beyond the next token, suffering from a snowballing of errors and struggling to predict the consequence of their actions several steps ahead. In this paper, we show how language models can be taught to search by representing the process of search in language, as a flattened string -- a stream of search (SoS). We propose a unified language for search that captures an array of different symbolic search strategies. We demonstrate our approach using the simple yet difficult game of Countdown, where the go"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03683","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/2404.03683/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":"2404.03683","created_at":"2026-07-05T08:04:39.967921+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.03683v1","created_at":"2026-07-05T08:04:39.967921+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03683","created_at":"2026-07-05T08:04:39.967921+00:00"},{"alias_kind":"pith_short_12","alias_value":"TTUABZN2SB4C","created_at":"2026-07-05T08:04:39.967921+00:00"},{"alias_kind":"pith_short_16","alias_value":"TTUABZN2SB4C5J2K","created_at":"2026-07-05T08:04:39.967921+00:00"},{"alias_kind":"pith_short_8","alias_value":"TTUABZN2","created_at":"2026-07-05T08:04:39.967921+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07492","citing_title":"Search, Fail, Recover: A Training Framework for Correction-Aware Reasoning","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2606.18022","citing_title":"Recursive Scaling in Masked Diffusion Models","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02234","citing_title":"Purified OPSD: On-Policy Self-Distillation Without Losing How to Think","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27570","citing_title":"LaneRoPE: Positional Encoding for Collaborative Parallel Reasoning and Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29123","citing_title":"The Confidence Shortcut: A Reasoning Failure Mode of Masked Diffusion Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06556","citing_title":"Robots Need More than VLA and World Models","ref_index":148,"is_internal_anchor":false},{"citing_arxiv_id":"2501.19201","citing_title":"Efficient Reasoning with Hidden Thinking","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2503.22693","citing_title":"Bridging Language Models and Financial Analysis","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09732","citing_title":"Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2408.07199","citing_title":"Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19944","citing_title":"A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":127,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04476","citing_title":"Vision-aligned Latent Reasoning for Multi-modal Large Language Model","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2412.18925","citing_title":"HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2502.17419","citing_title":"From System 1 to System 2: A Survey of Reasoning Large Language Models","ref_index":255,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08221","citing_title":"NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01373","citing_title":"Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrast","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19567","citing_title":"Multi-modal Reasoning with LLMs for Visual Semantic Arithmetic","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06769","citing_title":"Training Large Language Models to Reason in a Continuous Latent Space","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU","json":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU.json","graph_json":"https://pith.science/api/pith-number/TTUABZN2SB4C5J2KCXUMQNEEGU/graph.json","events_json":"https://pith.science/api/pith-number/TTUABZN2SB4C5J2KCXUMQNEEGU/events.json","paper":"https://pith.science/paper/TTUABZN2"},"agent_actions":{"view_html":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU","download_json":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU.json","view_paper":"https://pith.science/paper/TTUABZN2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.03683&json=true","fetch_graph":"https://pith.science/api/pith-number/TTUABZN2SB4C5J2KCXUMQNEEGU/graph.json","fetch_events":"https://pith.science/api/pith-number/TTUABZN2SB4C5J2KCXUMQNEEGU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU/action/storage_attestation","attest_author":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU/action/author_attestation","sign_citation":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU/action/citation_signature","submit_replication":"https://pith.science/pith/TTUABZN2SB4C5J2KCXUMQNEEGU/action/replication_record"}},"created_at":"2026-07-05T08:04:39.967921+00:00","updated_at":"2026-07-05T08:04:39.967921+00:00"}