{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PLZXNFTJQXV5FO6CYZ7YVGIZYS","short_pith_number":"pith:PLZXNFTJ","schema_version":"1.0","canonical_sha256":"7af376966985ebd2bbc2c67f8a9919c4a803cbb292f7fa2711b508de74bc7a23","source":{"kind":"arxiv","id":"2506.08295","version":1},"attestation_state":"computed","paper":{"title":"From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Han, Jiangchao Yao, Sanmi Koyejo, Xiao Feng, Zhanke Zhou, Zhaocheng Zhu","submitted_at":"2025-06-09T23:56:41Z","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"},"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":"2506.08295","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-09T23:56:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"7480462234f06b874c0f103fa4b0159a7ac9ce5641b8ad1028b7fa53ee1c7c08","abstract_canon_sha256":"3d4ecb19c1485110002eb6ee12e8aaae3ff1ea3d916f136c0452b9ba943be48f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:09.107308Z","signature_b64":"eVA0LC8ocgHjWpKh+HcqyKUIpQgZmQG+GLX8c5KaxMkHoz4zKXLqCMCiSJyebCmn+F887xeLSX3itu78Hc07AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7af376966985ebd2bbc2c67f8a9919c4a803cbb292f7fa2711b508de74bc7a23","last_reissued_at":"2026-07-05T11:19:09.106890Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:09.106890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Passive to Active Reasoning: Can Large Language Models Ask the Right Questions under Incomplete Information?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bo Han, Jiangchao Yao, Sanmi Koyejo, Xiao Feng, Zhanke Zhou, Zhaocheng Zhu","submitted_at":"2025-06-09T23:56:41Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08295","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/2506.08295/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":"2506.08295","created_at":"2026-07-05T11:19:09.106940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.08295v1","created_at":"2026-07-05T11:19:09.106940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08295","created_at":"2026-07-05T11:19:09.106940+00:00"},{"alias_kind":"pith_short_12","alias_value":"PLZXNFTJQXV5","created_at":"2026-07-05T11:19:09.106940+00:00"},{"alias_kind":"pith_short_16","alias_value":"PLZXNFTJQXV5FO6C","created_at":"2026-07-05T11:19:09.106940+00:00"},{"alias_kind":"pith_short_8","alias_value":"PLZXNFTJ","created_at":"2026-07-05T11:19:09.106940+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08068","citing_title":"DICE: Entropy-Regularized Equilibrium Selection for Stable Multi-Agent LLM Coordination","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03135","citing_title":"Uncertainty-Aware Clarification in LLM Agents with Information Gain","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01182","citing_title":"CA-BED: Conversation-Aware Bayesian Experimental Design","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15207","citing_title":"TeamTR: Trust-Region Fine-Tuning for Multi-Agent LLM Coordination","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05216","citing_title":"SAT: Sequential Agent Tuning for Coordinator Free Plug and Play Multi-LLM Training with Monotonic Improvement Guarantees","ref_index":41,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS","json":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS.json","graph_json":"https://pith.science/api/pith-number/PLZXNFTJQXV5FO6CYZ7YVGIZYS/graph.json","events_json":"https://pith.science/api/pith-number/PLZXNFTJQXV5FO6CYZ7YVGIZYS/events.json","paper":"https://pith.science/paper/PLZXNFTJ"},"agent_actions":{"view_html":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS","download_json":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS.json","view_paper":"https://pith.science/paper/PLZXNFTJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.08295&json=true","fetch_graph":"https://pith.science/api/pith-number/PLZXNFTJQXV5FO6CYZ7YVGIZYS/graph.json","fetch_events":"https://pith.science/api/pith-number/PLZXNFTJQXV5FO6CYZ7YVGIZYS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS/action/storage_attestation","attest_author":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS/action/author_attestation","sign_citation":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS/action/citation_signature","submit_replication":"https://pith.science/pith/PLZXNFTJQXV5FO6CYZ7YVGIZYS/action/replication_record"}},"created_at":"2026-07-05T11:19:09.106940+00:00","updated_at":"2026-07-05T11:19:09.106940+00:00"}