{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PEW4HTIJ6O6GYIC5M665N3X64C","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"414129eaec3205a5b78792f69d5ce6a70187b46b7de1c03d7aa64f109913f380","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-29T16:55:32Z","title_canon_sha256":"a0d1742a2645f447c0c0bdacdd4d3b240299aa0143dd9885a215f781682917de"},"schema_version":"1.0","source":{"id":"2405.19262","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.19262","created_at":"2026-07-05T09:37:37Z"},{"alias_kind":"arxiv_version","alias_value":"2405.19262v3","created_at":"2026-07-05T09:37:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.19262","created_at":"2026-07-05T09:37:37Z"},{"alias_kind":"pith_short_12","alias_value":"PEW4HTIJ6O6G","created_at":"2026-07-05T09:37:37Z"},{"alias_kind":"pith_short_16","alias_value":"PEW4HTIJ6O6GYIC5","created_at":"2026-07-05T09:37:37Z"},{"alias_kind":"pith_short_8","alias_value":"PEW4HTIJ","created_at":"2026-07-05T09:37:37Z"}],"graph_snapshots":[{"event_id":"sha256:1016c0e062ce0d8e7486c810b808fc6007980b19a34db00f36bc05a07544582b","target":"graph","created_at":"2026-07-05T09:37:37Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2405.19262/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models are usually fine-tuned to align with human preferences. However, fine-tuning a large language model can be challenging. In this work, we introduce $\\textit{weak-to-strong search}$, framing the alignment of a large language model as a test-time greedy search to maximize the log-probability difference between small tuned and untuned models while sampling from the frozen large model. This method serves both as (1) a compute-efficient model up-scaling strategy that avoids directly tuning the large model and as (2) an instance of weak-to-strong generalization that enhances a s","authors_text":"Chao Yang, Jie Liu, Yu Qiao, Zhanhui Zhou, Zhichen Dong, Zhixuan Liu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-29T16:55:32Z","title":"Weak-to-Strong Search: Align Large Language Models via Searching over Small Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.19262","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:396430aeef4729e66153ff05f61470eb45ab7bb3783420d8cca375beb3e0a0cb","target":"record","created_at":"2026-07-05T09:37:37Z","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":"414129eaec3205a5b78792f69d5ce6a70187b46b7de1c03d7aa64f109913f380","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-29T16:55:32Z","title_canon_sha256":"a0d1742a2645f447c0c0bdacdd4d3b240299aa0143dd9885a215f781682917de"},"schema_version":"1.0","source":{"id":"2405.19262","kind":"arxiv","version":3}},"canonical_sha256":"792dc3cd09f3bc6c205d67bdd6eefee0aa9a9f34d9a8856a5ff9c8b59ff7997d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"792dc3cd09f3bc6c205d67bdd6eefee0aa9a9f34d9a8856a5ff9c8b59ff7997d","first_computed_at":"2026-07-05T09:37:37.683780Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:37:37.683780Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fS/hfDXXSQKmXpBfYOkVC2yuTSTfbWNtWgOTlWKHgKTy24k+B0ifWPOOBX+TAIcKcFp1Ob2l4XrQrZ/gN7knCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:37:37.684577Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.19262","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:396430aeef4729e66153ff05f61470eb45ab7bb3783420d8cca375beb3e0a0cb","sha256:1016c0e062ce0d8e7486c810b808fc6007980b19a34db00f36bc05a07544582b"],"state_sha256":"2a9d6ce69d03e111a6867952467aecf2120a753db8b5bf527fa120fcf67fe142"}