{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VXIL36EMYSK3K2WJIIUGZN4SHX","short_pith_number":"pith:VXIL36EM","schema_version":"1.0","canonical_sha256":"add0bdf88cc495b56ac942286cb7923de279a6b783700348972c1d7085c566a9","source":{"kind":"arxiv","id":"2212.10001","version":2},"attestation_state":"computed","paper":{"title":"Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boshi Wang, Huan Sun, Jiaming Shen, Luke Zettlemoyer, Sewon Min, Xiang Deng, You Wu","submitted_at":"2022-12-20T05:20:54Z","abstract_excerpt":"Chain-of-Thought (CoT) prompting can dramatically improve the multi-step reasoning abilities of large language models (LLMs). CoT explicitly encourages the LLM to generate intermediate rationales for solving a problem, by providing a series of reasoning steps in the demonstrations. Despite its success, there is still little understanding of what makes CoT prompting effective and which aspects of the demonstrated reasoning steps contribute to its performance. In this paper, we show that CoT reasoning is possible even with invalid demonstrations - prompting with invalid reasoning steps can achie"},"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":"2212.10001","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-12-20T05:20:54Z","cross_cats_sorted":[],"title_canon_sha256":"b3aa09ed89448e002938f12008873599066d6b39683cff76a65f3a06ddfd0e2b","abstract_canon_sha256":"cde01cea3c1cf7e1e71f138477cdb06fe3aad60c0c6a9a070856e71ecedb09c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:16:14.557135Z","signature_b64":"9ssbkD86LpUpUpHPMTqDUsQeEE3oTW7/p90hBv1yi2JAeuq9k5RxhySjKFg++z0d8dDSfiaI7e3JMFMiFrkYDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"add0bdf88cc495b56ac942286cb7923de279a6b783700348972c1d7085c566a9","last_reissued_at":"2026-07-05T06:16:14.556586Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:16:14.556586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boshi Wang, Huan Sun, Jiaming Shen, Luke Zettlemoyer, Sewon Min, Xiang Deng, You Wu","submitted_at":"2022-12-20T05:20:54Z","abstract_excerpt":"Chain-of-Thought (CoT) prompting can dramatically improve the multi-step reasoning abilities of large language models (LLMs). CoT explicitly encourages the LLM to generate intermediate rationales for solving a problem, by providing a series of reasoning steps in the demonstrations. Despite its success, there is still little understanding of what makes CoT prompting effective and which aspects of the demonstrated reasoning steps contribute to its performance. In this paper, we show that CoT reasoning is possible even with invalid demonstrations - prompting with invalid reasoning steps can achie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.10001","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/2212.10001/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":"2212.10001","created_at":"2026-07-05T06:16:14.556657+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.10001v2","created_at":"2026-07-05T06:16:14.556657+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.10001","created_at":"2026-07-05T06:16:14.556657+00:00"},{"alias_kind":"pith_short_12","alias_value":"VXIL36EMYSK3","created_at":"2026-07-05T06:16:14.556657+00:00"},{"alias_kind":"pith_short_16","alias_value":"VXIL36EMYSK3K2WJ","created_at":"2026-07-05T06:16:14.556657+00:00"},{"alias_kind":"pith_short_8","alias_value":"VXIL36EM","created_at":"2026-07-05T06:16:14.556657+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08068","citing_title":"DICE: Entropy-Regularized Equilibrium Selection for Stable Multi-Agent LLM Coordination","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2305.09617","citing_title":"Towards Expert-Level Medical Question Answering with Large Language Models","ref_index":99,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21035","citing_title":"GenoMAS: A Multi-Agent Framework for Scientific Discovery via Code-Driven Gene Expression Analysis","ref_index":117,"is_internal_anchor":false},{"citing_arxiv_id":"2409.12917","citing_title":"Training Language Models to Self-Correct via Reinforcement Learning","ref_index":125,"is_internal_anchor":false},{"citing_arxiv_id":"2309.00267","citing_title":"RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08221","citing_title":"NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2502.18864","citing_title":"Towards an AI co-scientist","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06769","citing_title":"Training Large Language Models to Reason in a Continuous Latent Space","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX","json":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX.json","graph_json":"https://pith.science/api/pith-number/VXIL36EMYSK3K2WJIIUGZN4SHX/graph.json","events_json":"https://pith.science/api/pith-number/VXIL36EMYSK3K2WJIIUGZN4SHX/events.json","paper":"https://pith.science/paper/VXIL36EM"},"agent_actions":{"view_html":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX","download_json":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX.json","view_paper":"https://pith.science/paper/VXIL36EM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.10001&json=true","fetch_graph":"https://pith.science/api/pith-number/VXIL36EMYSK3K2WJIIUGZN4SHX/graph.json","fetch_events":"https://pith.science/api/pith-number/VXIL36EMYSK3K2WJIIUGZN4SHX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX/action/storage_attestation","attest_author":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX/action/author_attestation","sign_citation":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX/action/citation_signature","submit_replication":"https://pith.science/pith/VXIL36EMYSK3K2WJIIUGZN4SHX/action/replication_record"}},"created_at":"2026-07-05T06:16:14.556657+00:00","updated_at":"2026-07-05T06:16:14.556657+00:00"}