{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PSBDAAIW4NITRKXQLLFT2ZUQRN","short_pith_number":"pith:PSBDAAIW","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"},"canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","source":{"kind":"arxiv","id":"2505.11556","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11556","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11556v4","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11556","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"PSBDAAIW4NIT","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"PSBDAAIW4NITRKXQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"PSBDAAIW","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PSBDAAIW4NITRKXQLLFT2ZUQRN","target":"record","payload":{"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"},"canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","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"},"source_kind":"arxiv","source_id":"2505.11556","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nUr+/b+8FSjPP721Q3hn8CyrUeBksHYf+33lJr/EYgC8sAi6ZXBt34VHGXs0diFbFAzSg5+OS/eYsOEVCN7dBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T12:05:24.343854Z"},"content_sha256":"185fe61ebd0da8eb06e5f282ed63109eb762e491ced22c614c767561493a4f54","schema_version":"1.0","event_id":"sha256:185fe61ebd0da8eb06e5f282ed63109eb762e491ced22c614c767561493a4f54"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PSBDAAIW4NITRKXQLLFT2ZUQRN","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rWX78iH6BgpUP7Rwi5NKvYRRTT2iMPISCg6N+EamQSxAbTH8zcM2CpRhJI8qbEqyB4MFlgzrXPDmCvUItGwBBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T12:05:24.344195Z"},"content_sha256":"edcf484d7de9193ec7b6a1eb217a1db7c770d47793c921ab61cf043added15f2","schema_version":"1.0","event_id":"sha256:edcf484d7de9193ec7b6a1eb217a1db7c770d47793c921ab61cf043added15f2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/bundle.json","state_url":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-30T12:05:24Z","links":{"resolver":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN","bundle":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/bundle.json","state":"https://pith.science/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PSBDAAIW4NITRKXQLLFT2ZUQRN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PSBDAAIW4NITRKXQLLFT2ZUQRN","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"77872c9885ea41df4ea3a9d19c438c09e7dfc83bec1f0c08e6fd0f4c34fbb127","cross_cats_sorted":["cs.AI","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T19:22:54Z","title_canon_sha256":"72203059e684bb774410f72e73a4fd8fe512bd253da6773ed96bc9133875db7a"},"schema_version":"1.0","source":{"id":"2505.11556","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11556","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11556v4","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11556","created_at":"2026-05-18T03:09:35Z"},{"alias_kind":"pith_short_12","alias_value":"PSBDAAIW4NIT","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"PSBDAAIW4NITRKXQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"PSBDAAIW","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:edcf484d7de9193ec7b6a1eb217a1db7c770d47793c921ab61cf043added15f2","target":"graph","created_at":"2026-05-18T03:09:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"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","authors_text":"Aoi Naito, Hirokazu Shirado, Yuxuan Li","cross_cats":["cs.AI","cs.MA"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T19:22:54Z","title":"Systematic Failures in Collective Reasoning under Distributed Information in Multi-Agent LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11556","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:185fe61ebd0da8eb06e5f282ed63109eb762e491ced22c614c767561493a4f54","target":"record","created_at":"2026-05-18T03:09:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"77872c9885ea41df4ea3a9d19c438c09e7dfc83bec1f0c08e6fd0f4c34fbb127","cross_cats_sorted":["cs.AI","cs.MA"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T19:22:54Z","title_canon_sha256":"72203059e684bb774410f72e73a4fd8fe512bd253da6773ed96bc9133875db7a"},"schema_version":"1.0","source":{"id":"2505.11556","kind":"arxiv","version":4}},"canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7c82300116e35138aaf05acb3d66908b57cd46c3f5ff303e1522f0968449e660","first_computed_at":"2026-05-18T03:09:35.742120Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:35.742120Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xIw7N3g5yFUtp99P+sh2HjpSWgrbFmB/ak/aARedr+Q5/RJ03LcCNZIr3wHR4mOCrn9BOiUKRwtDbHkjVQM4AQ==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:35.742658Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11556","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:185fe61ebd0da8eb06e5f282ed63109eb762e491ced22c614c767561493a4f54","sha256:edcf484d7de9193ec7b6a1eb217a1db7c770d47793c921ab61cf043added15f2"],"state_sha256":"40b1c1bf98ee559545369df8216375961361fc5843334d28697adfaf5a3cfc5f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"84kkdUaBl0Fpe9EZ8IY8HTQJF8Jo+dudg1QBpmPZIZbuRT9+EJB6eIdTiL7HBUjQv+m2cJZAUxldhJGlF5VlAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T12:05:24.346294Z","bundle_sha256":"baeaf6ab53f26a94ffc48f848267920791d453a4caf869d5d3924a86d864d71e"}}