{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ANJLWVU3ITETA4TQ4N7WZKEOVE","short_pith_number":"pith:ANJLWVU3","schema_version":"1.0","canonical_sha256":"0352bb569b44c9307270e37f6ca88ea929f0a623110384bcafc035bec5fc815f","source":{"kind":"arxiv","id":"2311.17371","version":3},"attestation_state":"computed","paper":{"title":"Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Andries Smit, Arnu Pretorius, Nathan Grinsztajn, Paul Duckworth, Thomas D. Barrett","submitted_at":"2023-11-29T05:54:41Z","abstract_excerpt":"Recent advancements in large language models (LLMs) underscore their potential for responding to inquiries in various domains. However, ensuring that generative agents provide accurate and reliable answers remains an ongoing challenge. In this context, multi-agent debate (MAD) has emerged as a promising strategy for enhancing the truthfulness of LLMs. We benchmark a range of debating and prompting strategies to explore the trade-offs between cost, time, and accuracy. Importantly, we find that multi-agent debating systems, in their current form, do not reliably outperform other proposed prompti"},"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":"2311.17371","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-29T05:54:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9187be730c7194e7873f35430a2b7f3d0141912e7c64defe9e25fd0def5560d4","abstract_canon_sha256":"58df7925302f82a5ab50302d214440720abe36769116bac629bedc4345b1b4e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:45:23.611547Z","signature_b64":"5l4g5b3tNCfFfa9wvv0X2pwBIQ1xsn9D5Nnz/8kN0J9lA4eZBKzc1FDnETIv0qHhHHh9QLO0X8Fn1+vWl1/TDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0352bb569b44c9307270e37f6ca88ea929f0a623110384bcafc035bec5fc815f","last_reissued_at":"2026-07-05T08:45:23.611046Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:45:23.611046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Andries Smit, Arnu Pretorius, Nathan Grinsztajn, Paul Duckworth, Thomas D. Barrett","submitted_at":"2023-11-29T05:54:41Z","abstract_excerpt":"Recent advancements in large language models (LLMs) underscore their potential for responding to inquiries in various domains. However, ensuring that generative agents provide accurate and reliable answers remains an ongoing challenge. In this context, multi-agent debate (MAD) has emerged as a promising strategy for enhancing the truthfulness of LLMs. We benchmark a range of debating and prompting strategies to explore the trade-offs between cost, time, and accuracy. Importantly, we find that multi-agent debating systems, in their current form, do not reliably outperform other proposed prompti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.17371","kind":"arxiv","version":3},"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/2311.17371/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":"2311.17371","created_at":"2026-07-05T08:45:23.611107+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.17371v3","created_at":"2026-07-05T08:45:23.611107+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.17371","created_at":"2026-07-05T08:45:23.611107+00:00"},{"alias_kind":"pith_short_12","alias_value":"ANJLWVU3ITET","created_at":"2026-07-05T08:45:23.611107+00:00"},{"alias_kind":"pith_short_16","alias_value":"ANJLWVU3ITETA4TQ","created_at":"2026-07-05T08:45:23.611107+00:00"},{"alias_kind":"pith_short_8","alias_value":"ANJLWVU3","created_at":"2026-07-05T08:45:23.611107+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25191","citing_title":"To Isolate or to Score? Model-Adaptive Assessment for Cost-Efficient Multi-Agent RAG","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22949","citing_title":"MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22949","citing_title":"MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2601.22297","citing_title":"Learning from Self-Debate: Preparing Reasoning Models for Multi-Agent Debate","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03809","citing_title":"Representational Collapse in Multi-Agent LLM Committees: Measurement and Diversity-Aware Consensus","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10528","citing_title":"Collective Alignment in LLM Multi-Agent Systems: Disentangling Bias from Cooperation via Statistical Physics","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09278","citing_title":"EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibrium","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10389","citing_title":"BLUEmed: Retrieval-Augmented Multi-Agent Debate for Clinical Error Detection","ref_index":46,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE","json":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE.json","graph_json":"https://pith.science/api/pith-number/ANJLWVU3ITETA4TQ4N7WZKEOVE/graph.json","events_json":"https://pith.science/api/pith-number/ANJLWVU3ITETA4TQ4N7WZKEOVE/events.json","paper":"https://pith.science/paper/ANJLWVU3"},"agent_actions":{"view_html":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE","download_json":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE.json","view_paper":"https://pith.science/paper/ANJLWVU3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.17371&json=true","fetch_graph":"https://pith.science/api/pith-number/ANJLWVU3ITETA4TQ4N7WZKEOVE/graph.json","fetch_events":"https://pith.science/api/pith-number/ANJLWVU3ITETA4TQ4N7WZKEOVE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE/action/storage_attestation","attest_author":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE/action/author_attestation","sign_citation":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE/action/citation_signature","submit_replication":"https://pith.science/pith/ANJLWVU3ITETA4TQ4N7WZKEOVE/action/replication_record"}},"created_at":"2026-07-05T08:45:23.611107+00:00","updated_at":"2026-07-05T08:45:23.611107+00:00"}