{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2DDEPRRVC5S75SUXF52CJK5K7N","short_pith_number":"pith:2DDEPRRV","schema_version":"1.0","canonical_sha256":"d0c647c6351765feca972f7424abaafb4a7a6f99595d8589eaa0b2075322f3c9","source":{"kind":"arxiv","id":"2212.09561","version":5},"attestation_state":"computed","paper":{"title":"Large Language Models are Better Reasoners with Self-Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Bin Li, Bin Sun, Fei Xia, Jun Zhao, Kang Liu, Minjun Zhu, Shengping Liu, Shizhu He, Yixuan Weng","submitted_at":"2022-12-19T15:51:52Z","abstract_excerpt":"Recently, with the chain of thought (CoT) prompting, large language models (LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural language processing tasks such as arithmetic, commonsense, and logical reasoning. However, LLMs with CoT require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes and vulnerable to error accumulation. The above issues make the LLMs need the ability to verify the answers. In fact, after inferring conclusions in some thinking decision tasks, people often check them by re-verifying steps to avoid some"},"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":"2212.09561","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2022-12-19T15:51:52Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5c4aed317384593ce73f744eb0ae2516661d2be9e06ec107d6a05d19ace75d03","abstract_canon_sha256":"d97dd45183ad483527870e8b1d4d2ab0cd3a89fe55b562af9daf3fc50d1d1c10"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:26.633721Z","signature_b64":"1Tk1KlmE1y9IUQTFAHvpF+XtKCUSJjYQsKeBL+S1r2WhUAOZ5skbGbRVveTtQTxgY6qh6WE9uJV0LF6SJtFsAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0c647c6351765feca972f7424abaafb4a7a6f99595d8589eaa0b2075322f3c9","last_reissued_at":"2026-07-05T07:02:26.633283Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:26.633283Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models are Better Reasoners with Self-Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Bin Li, Bin Sun, Fei Xia, Jun Zhao, Kang Liu, Minjun Zhu, Shengping Liu, Shizhu He, Yixuan Weng","submitted_at":"2022-12-19T15:51:52Z","abstract_excerpt":"Recently, with the chain of thought (CoT) prompting, large language models (LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural language processing tasks such as arithmetic, commonsense, and logical reasoning. However, LLMs with CoT require multi-step prompting and multi-token prediction, which is highly sensitive to individual mistakes and vulnerable to error accumulation. The above issues make the LLMs need the ability to verify the answers. In fact, after inferring conclusions in some thinking decision tasks, people often check them by re-verifying steps to avoid some"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09561","kind":"arxiv","version":5},"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/2212.09561/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":"2212.09561","created_at":"2026-07-05T07:02:26.633355+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.09561v5","created_at":"2026-07-05T07:02:26.633355+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09561","created_at":"2026-07-05T07:02:26.633355+00:00"},{"alias_kind":"pith_short_12","alias_value":"2DDEPRRVC5S7","created_at":"2026-07-05T07:02:26.633355+00:00"},{"alias_kind":"pith_short_16","alias_value":"2DDEPRRVC5S75SUX","created_at":"2026-07-05T07:02:26.633355+00:00"},{"alias_kind":"pith_short_8","alias_value":"2DDEPRRV","created_at":"2026-07-05T07:02:26.633355+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05391","citing_title":"LLM-as-a-Verifier: A General-Purpose Verification Framework","ref_index":35,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":259,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00172","citing_title":"CAST: Non-Privileged Clipped Asymmetric Self-Teaching with Advantage Flipping for GRPO","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":258,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24790","citing_title":"Grad Detect: Gradient-Based Hallucination Detection in LLMs","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2411.18104","citing_title":"Training and Evaluating Language Models with Template-based Data Generation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2309.11495","citing_title":"Chain-of-Verification Reduces Hallucination in Large Language Models","ref_index":166,"is_internal_anchor":false},{"citing_arxiv_id":"2512.00127","citing_title":"Generating Verifiable Chain of Thoughts from Exection-Traces","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2503.10615","citing_title":"R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08904","citing_title":"OPT-BENCH: Evaluating the Iterative Self-Optimization of LLM Agents in Large-Scale Search Spaces","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N","json":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N.json","graph_json":"https://pith.science/api/pith-number/2DDEPRRVC5S75SUXF52CJK5K7N/graph.json","events_json":"https://pith.science/api/pith-number/2DDEPRRVC5S75SUXF52CJK5K7N/events.json","paper":"https://pith.science/paper/2DDEPRRV"},"agent_actions":{"view_html":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N","download_json":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N.json","view_paper":"https://pith.science/paper/2DDEPRRV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.09561&json=true","fetch_graph":"https://pith.science/api/pith-number/2DDEPRRVC5S75SUXF52CJK5K7N/graph.json","fetch_events":"https://pith.science/api/pith-number/2DDEPRRVC5S75SUXF52CJK5K7N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N/action/storage_attestation","attest_author":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N/action/author_attestation","sign_citation":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N/action/citation_signature","submit_replication":"https://pith.science/pith/2DDEPRRVC5S75SUXF52CJK5K7N/action/replication_record"}},"created_at":"2026-07-05T07:02:26.633355+00:00","updated_at":"2026-07-05T07:02:26.633355+00:00"}