{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z6EKFBPS44TTEM36P5LKPDNLQZ","short_pith_number":"pith:Z6EKFBPS","schema_version":"1.0","canonical_sha256":"cf88a285f2e72732337e7f56a78dab86460a824af98562f4ab55a48604e348a8","source":{"kind":"arxiv","id":"2505.22960","version":2},"attestation_state":"computed","paper":{"title":"Revisiting Multi-Agent Debate as Test-Time Scaling: A Systematic Study of Conditional Effectiveness","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Euiin Yi, Jongwoo Ko, Kimin Lee, Se-Young Yun, Yongjin Yang, Zhijing Jin","submitted_at":"2025-05-29T01:02:55Z","abstract_excerpt":"The remarkable growth in large language model (LLM) capabilities has spurred exploration into multi-agent systems, with debate frameworks emerging as a promising avenue for enhanced problem-solving. These multi-agent debate (MAD) approaches, where agents collaboratively present, critique, and refine arguments, potentially offer improved reasoning, robustness, and diverse perspectives over monolithic models. Despite prior studies leveraging MAD, a systematic understanding of its effectiveness compared to self-agent methods, particularly under varying conditions, remains elusive. This paper seek"},"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":"2505.22960","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-29T01:02:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"f3584421625ceefb9f9e96ad95b0f67cba93d7c506f537b576b03a08a51b35a5","abstract_canon_sha256":"2b0874417e46da519d164e9b8a29f8618b6cd5c87cab2953fc93586813fa2605"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:21.396100Z","signature_b64":"6nmFwycmQVBDCIFlT8z3ol1FI7hGCLkfRUmugezSDlAL0BGpDL9xjb1cYWtbrcC2bJlin46axbqwNZlG5oJfAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf88a285f2e72732337e7f56a78dab86460a824af98562f4ab55a48604e348a8","last_reissued_at":"2026-07-05T11:24:21.395563Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:21.395563Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Revisiting Multi-Agent Debate as Test-Time Scaling: A Systematic Study of Conditional Effectiveness","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Euiin Yi, Jongwoo Ko, Kimin Lee, Se-Young Yun, Yongjin Yang, Zhijing Jin","submitted_at":"2025-05-29T01:02:55Z","abstract_excerpt":"The remarkable growth in large language model (LLM) capabilities has spurred exploration into multi-agent systems, with debate frameworks emerging as a promising avenue for enhanced problem-solving. These multi-agent debate (MAD) approaches, where agents collaboratively present, critique, and refine arguments, potentially offer improved reasoning, robustness, and diverse perspectives over monolithic models. Despite prior studies leveraging MAD, a systematic understanding of its effectiveness compared to self-agent methods, particularly under varying conditions, remains elusive. This paper seek"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22960","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/2505.22960/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":"2505.22960","created_at":"2026-07-05T11:24:21.395630+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.22960v2","created_at":"2026-07-05T11:24:21.395630+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22960","created_at":"2026-07-05T11:24:21.395630+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z6EKFBPS44TT","created_at":"2026-07-05T11:24:21.395630+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z6EKFBPS44TTEM36","created_at":"2026-07-05T11:24:21.395630+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z6EKFBPS","created_at":"2026-07-05T11:24:21.395630+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19826","citing_title":"Heterogeneous LLM Debate Under Adversarial Peers: Honest Gains, Replacement Costs, and Resilience","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02866","citing_title":"When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02646","citing_title":"The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00820","citing_title":"Not All Flips Are Conformity: Decomposing Stance Convergence in Multi-Agent LLM Debate","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01704","citing_title":"The Reasoning Trap: An Information-Theoretic Bound on Closed-System Multi-Step LLM Reasoning","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ","json":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ.json","graph_json":"https://pith.science/api/pith-number/Z6EKFBPS44TTEM36P5LKPDNLQZ/graph.json","events_json":"https://pith.science/api/pith-number/Z6EKFBPS44TTEM36P5LKPDNLQZ/events.json","paper":"https://pith.science/paper/Z6EKFBPS"},"agent_actions":{"view_html":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ","download_json":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ.json","view_paper":"https://pith.science/paper/Z6EKFBPS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.22960&json=true","fetch_graph":"https://pith.science/api/pith-number/Z6EKFBPS44TTEM36P5LKPDNLQZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Z6EKFBPS44TTEM36P5LKPDNLQZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ/action/storage_attestation","attest_author":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ/action/author_attestation","sign_citation":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ/action/citation_signature","submit_replication":"https://pith.science/pith/Z6EKFBPS44TTEM36P5LKPDNLQZ/action/replication_record"}},"created_at":"2026-07-05T11:24:21.395630+00:00","updated_at":"2026-07-05T11:24:21.395630+00:00"}