{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:U4ID6NJRNXWAHCLLRAKS4T5ZOX","short_pith_number":"pith:U4ID6NJR","schema_version":"1.0","canonical_sha256":"a7103f35316dec03896b88152e4fb975f897cd9cb80b421a6c67d9d5b8bac513","source":{"kind":"arxiv","id":"2503.01935","version":1},"attestation_state":"computed","paper":{"title":"MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CY"],"primary_cat":"cs.MA","authors_text":"Cheng Qian, Heng Ji, Hongyi Du, Jiaxuan You, Kunlun Zhu, Shuyi Guo, Xiangru Tang, Xiaocheng Yang, Zhaochen Hong, Zhenhailong Wang, Zhe Wang","submitted_at":"2025-03-03T05:18:50Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents, yet existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition. In this paper, we introduce MultiAgentBench, a comprehensive benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios. Our framework measures not only task completion but also the quality of collaboration and competition using novel, milestone-based key performance indicators. Moreover, we evaluate v"},"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":"2503.01935","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2025-03-03T05:18:50Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CY"],"title_canon_sha256":"ce39eb13710a9858f5578458f6810b36906857eb3725b8c84ec7eb717b90025b","abstract_canon_sha256":"0c97c0d71461757cbc61049490d0950b2ffa952a7ea4cc2424c40fb093deed1d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:23:32.885510Z","signature_b64":"xwFF8//dnyxbEllOQYHh4sHDqj72Rm2Kcj9YlpHY5yx6A739JXzYh39K4sB+H9sZuvVJR86N8AJs2ppL+uF7Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a7103f35316dec03896b88152e4fb975f897cd9cb80b421a6c67d9d5b8bac513","last_reissued_at":"2026-07-05T10:23:32.884895Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:23:32.884895Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CY"],"primary_cat":"cs.MA","authors_text":"Cheng Qian, Heng Ji, Hongyi Du, Jiaxuan You, Kunlun Zhu, Shuyi Guo, Xiangru Tang, Xiaocheng Yang, Zhaochen Hong, Zhenhailong Wang, Zhe Wang","submitted_at":"2025-03-03T05:18:50Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents, yet existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition. In this paper, we introduce MultiAgentBench, a comprehensive benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios. Our framework measures not only task completion but also the quality of collaboration and competition using novel, milestone-based key performance indicators. Moreover, we evaluate v"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01935","kind":"arxiv","version":1},"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/2503.01935/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":"2503.01935","created_at":"2026-07-05T10:23:32.884959+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.01935v1","created_at":"2026-07-05T10:23:32.884959+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01935","created_at":"2026-07-05T10:23:32.884959+00:00"},{"alias_kind":"pith_short_12","alias_value":"U4ID6NJRNXWA","created_at":"2026-07-05T10:23:32.884959+00:00"},{"alias_kind":"pith_short_16","alias_value":"U4ID6NJRNXWAHCLL","created_at":"2026-07-05T10:23:32.884959+00:00"},{"alias_kind":"pith_short_8","alias_value":"U4ID6NJR","created_at":"2026-07-05T10:23:32.884959+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09800","citing_title":"FASE: Fast Adaptive Semantic Entropy for Code Quality","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28425","citing_title":"Tool Use Enables Undetectable Steganography in Multi-Agent LLM Systems","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30546","citing_title":"MAS-Lab: A Specification-Driven Validation Framework for Reliable Multi-Agent Systems","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28655","citing_title":"AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20629","citing_title":"Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2504.19678","citing_title":"From LLM Reasoning to Autonomous AI Agents: A Comprehensive Review","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2511.20857","citing_title":"Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory","ref_index":91,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21046","citing_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01487","citing_title":"AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07982","citing_title":"$\\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08761","citing_title":"Beyond the All-in-One Agent: Benchmarking Role-Specialized Multi-Agent Collaboration in Enterprise Workflows","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18805","citing_title":"AI scientists produce results without reasoning scientifically","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07073","citing_title":"TeamBench: Evaluating Agent Coordination under Enforced Role Separation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07935","citing_title":"TraceFix: Repairing Agent Coordination Protocols with TLA+ Counterexamples","ref_index":45,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX","json":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX.json","graph_json":"https://pith.science/api/pith-number/U4ID6NJRNXWAHCLLRAKS4T5ZOX/graph.json","events_json":"https://pith.science/api/pith-number/U4ID6NJRNXWAHCLLRAKS4T5ZOX/events.json","paper":"https://pith.science/paper/U4ID6NJR"},"agent_actions":{"view_html":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX","download_json":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX.json","view_paper":"https://pith.science/paper/U4ID6NJR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.01935&json=true","fetch_graph":"https://pith.science/api/pith-number/U4ID6NJRNXWAHCLLRAKS4T5ZOX/graph.json","fetch_events":"https://pith.science/api/pith-number/U4ID6NJRNXWAHCLLRAKS4T5ZOX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX/action/storage_attestation","attest_author":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX/action/author_attestation","sign_citation":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX/action/citation_signature","submit_replication":"https://pith.science/pith/U4ID6NJRNXWAHCLLRAKS4T5ZOX/action/replication_record"}},"created_at":"2026-07-05T10:23:32.884959+00:00","updated_at":"2026-07-05T10:23:32.884959+00:00"}