{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PSBDAAIW4NITRKXQLLFT2ZUQRN","short_pith_number":"pith:PSBDAAIW","schema_version":"1.0","canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","source":{"kind":"arxiv","id":"2505.11556","version":4},"attestation_state":"computed","paper":{"title":"Systematic Failures in Collective Reasoning under Distributed Information in Multi-Agent LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.CL","authors_text":"Aoi Naito, Hirokazu Shirado, Yuxuan Li","submitted_at":"2025-05-15T19:22:54Z","abstract_excerpt":"Multi-agent systems built on large language models (LLMs) are expected to enhance decision-making by pooling distributed information, yet systematically evaluating this capability has remained challenging. We introduce HiddenBench, a 65-task benchmark grounded in the Hidden Profile paradigm, which isolates collective reasoning under distributed information from individual reasoning ability. Evaluating 15 frontier LLMs, we find that multi-agent LLMs achieve only 30.1% accuracy under distributed information, compared to 80.7% accuracy for single agents given complete information. We trace this g"},"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":"2505.11556","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T19:22:54Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"72203059e684bb774410f72e73a4fd8fe512bd253da6773ed96bc9133875db7a","abstract_canon_sha256":"77872c9885ea41df4ea3a9d19c438c09e7dfc83bec1f0c08e6fd0f4c34fbb127"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:35.742658Z","signature_b64":"xIw7N3g5yFUtp99P+sh2HjpSWgrbFmB/ak/aARedr+Q5/RJ03LcCNZIr3wHR4mOCrn9BOiUKRwtDbHkjVQM4AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","last_reissued_at":"2026-05-18T03:09:35.742120Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:35.742120Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Systematic Failures in Collective Reasoning under Distributed Information in Multi-Agent LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.CL","authors_text":"Aoi Naito, Hirokazu Shirado, Yuxuan Li","submitted_at":"2025-05-15T19:22:54Z","abstract_excerpt":"Multi-agent systems built on large language models (LLMs) are expected to enhance decision-making by pooling distributed information, yet systematically evaluating this capability has remained challenging. We introduce HiddenBench, a 65-task benchmark grounded in the Hidden Profile paradigm, which isolates collective reasoning under distributed information from individual reasoning ability. Evaluating 15 frontier LLMs, we find that multi-agent LLMs achieve only 30.1% accuracy under distributed information, compared to 80.7% accuracy for single agents given complete information. We trace this g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11556","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2505.11556","created_at":"2026-05-18T03:09:35.742204+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11556v4","created_at":"2026-05-18T03:09:35.742204+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11556","created_at":"2026-05-18T03:09:35.742204+00:00"},{"alias_kind":"pith_short_12","alias_value":"PSBDAAIW4NIT","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"PSBDAAIW4NITRKXQ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"PSBDAAIW","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2509.18052","citing_title":"The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2510.19973","citing_title":"A Tutorial on Cognitive Biases in Agentic AI-Driven 6G Autonomous Networks","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2511.19175","citing_title":"LLM-Based Agentic Negotiation for 6G: Addressing Uncertainty Neglect and Tail-Event Risk","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07073","citing_title":"TeamBench: Evaluating Agent Coordination under Enforced Role Separation","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN","json":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN.json","graph_json":"https://pith.science/api/pith-number/PSBDAAIW4NITRKXQLLFT2ZUQRN/graph.json","events_json":"https://pith.science/api/pith-number/PSBDAAIW4NITRKXQLLFT2ZUQRN/events.json","paper":"https://pith.science/paper/PSBDAAIW"},"agent_actions":{"view_html":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN","download_json":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN.json","view_paper":"https://pith.science/paper/PSBDAAIW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11556&json=true","fetch_graph":"https://pith.science/api/pith-number/PSBDAAIW4NITRKXQLLFT2ZUQRN/graph.json","fetch_events":"https://pith.science/api/pith-number/PSBDAAIW4NITRKXQLLFT2ZUQRN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/action/storage_attestation","attest_author":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/action/author_attestation","sign_citation":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/action/citation_signature","submit_replication":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/action/replication_record"}},"created_at":"2026-05-18T03:09:35.742204+00:00","updated_at":"2026-05-18T03:09:35.742204+00:00"}