{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PLZXNFTJQXV5FO6CYZ7YVGIZYS","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":"3d4ecb19c1485110002eb6ee12e8aaae3ff1ea3d916f136c0452b9ba943be48f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T23:56:41Z","title_canon_sha256":"7480462234f06b874c0f103fa4b0159a7ac9ce5641b8ad1028b7fa53ee1c7c08"},"schema_version":"1.0","source":{"id":"2506.08295","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08295","created_at":"2026-07-05T11:19:09Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08295v1","created_at":"2026-07-05T11:19:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08295","created_at":"2026-07-05T11:19:09Z"},{"alias_kind":"pith_short_12","alias_value":"PLZXNFTJQXV5","created_at":"2026-07-05T11:19:09Z"},{"alias_kind":"pith_short_16","alias_value":"PLZXNFTJQXV5FO6C","created_at":"2026-07-05T11:19:09Z"},{"alias_kind":"pith_short_8","alias_value":"PLZXNFTJ","created_at":"2026-07-05T11:19:09Z"}],"graph_snapshots":[{"event_id":"sha256:b68f7dfabb8ef8bba2df3b2395316cdf5f94f24df09adeaf11d259fb66cdc17a","target":"graph","created_at":"2026-07-05T11:19:09Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.08295/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While existing benchmarks probe the reasoning abilities of large language models (LLMs) across diverse domains, they predominantly assess passive reasoning, providing models with all the information needed to reach a solution. By contrast, active reasoning-where an LLM must interact with external systems to acquire missing evidence or data-has received little systematic attention. To address this shortfall, we present AR-Bench, a novel benchmark designed explicitly to evaluate an LLM's active reasoning skills. AR-Bench comprises three task families-detective cases, situation puzzles, and guess","authors_text":"Bo Han, Jiangchao Yao, Sanmi Koyejo, Xiao Feng, Zhanke Zhou, Zhaocheng Zhu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T23:56:41Z","title":"From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08295","kind":"arxiv","version":1},"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:f246556e492b3fc75476de18ffd689f466ce31a4da0456d56542a12eae724704","target":"record","created_at":"2026-07-05T11:19:09Z","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":"3d4ecb19c1485110002eb6ee12e8aaae3ff1ea3d916f136c0452b9ba943be48f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T23:56:41Z","title_canon_sha256":"7480462234f06b874c0f103fa4b0159a7ac9ce5641b8ad1028b7fa53ee1c7c08"},"schema_version":"1.0","source":{"id":"2506.08295","kind":"arxiv","version":1}},"canonical_sha256":"7af376966985ebd2bbc2c67f8a9919c4a803cbb292f7fa2711b508de74bc7a23","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7af376966985ebd2bbc2c67f8a9919c4a803cbb292f7fa2711b508de74bc7a23","first_computed_at":"2026-07-05T11:19:09.106890Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:09.106890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eVA0LC8ocgHjWpKh+HcqyKUIpQgZmQG+GLX8c5KaxMkHoz4zKXLqCMCiSJyebCmn+F887xeLSX3itu78Hc07AA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:09.107308Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.08295","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f246556e492b3fc75476de18ffd689f466ce31a4da0456d56542a12eae724704","sha256:b68f7dfabb8ef8bba2df3b2395316cdf5f94f24df09adeaf11d259fb66cdc17a"],"state_sha256":"c936e8a96dc7e446c7df8785526062774a8b7157cf12c65b2e7051d93cc79424"}