{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:T7RMH2T3SGRUZEDC3HW6TGPJRG","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":"593788663157b252b6a57a3d5b576199cf37090f2be2fba594d1cb18a1ff72a4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-04T06:47:58Z","title_canon_sha256":"f82e8407a9d9c13603e48e95a93a6aad9f964dd76551332776898f1e5c8de739"},"schema_version":"1.0","source":{"id":"2310.03051","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.03051","created_at":"2026-07-05T06:57:28Z"},{"alias_kind":"arxiv_version","alias_value":"2310.03051v1","created_at":"2026-07-05T06:57:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.03051","created_at":"2026-07-05T06:57:28Z"},{"alias_kind":"pith_short_12","alias_value":"T7RMH2T3SGRU","created_at":"2026-07-05T06:57:28Z"},{"alias_kind":"pith_short_16","alias_value":"T7RMH2T3SGRUZEDC","created_at":"2026-07-05T06:57:28Z"},{"alias_kind":"pith_short_8","alias_value":"T7RMH2T3","created_at":"2026-07-05T06:57:28Z"}],"graph_snapshots":[{"event_id":"sha256:52595757ababf388355db3ddddd4d16fe75b265db400e8fb31e90574285e773a","target":"graph","created_at":"2026-07-05T06:57:28Z","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/2310.03051/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"\"Thinking is for Doing.\" Humans can infer other people's mental states from observations--an ability called Theory-of-Mind (ToM)--and subsequently act pragmatically on those inferences. Existing question answering benchmarks such as ToMi ask models questions to make inferences about beliefs of characters in a story, but do not test whether models can then use these inferences to guide their actions. We propose a new evaluation paradigm for large language models (LLMs): Thinking for Doing (T4D), which requires models to connect inferences about others' mental states to actions in social scenari","authors_text":"Aditya Gupta, Aida Nematzadeh, Aman Madaan, Ari Holtzman, Jay Pujara, Kevin R. McKee, Manaal Faruqui, Pei Zhou, Shyam Upadhyay, Srividya Pranavi Potharaju, Swaroop Mishra, Xiang Ren","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-04T06:47:58Z","title":"How FaR Are Large Language Models From Agents with Theory-of-Mind?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.03051","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:b7fcb3214d832b8454213b14b0c9087e94b74c1ba39a9c1411ae7596a3499a9c","target":"record","created_at":"2026-07-05T06:57:28Z","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":"593788663157b252b6a57a3d5b576199cf37090f2be2fba594d1cb18a1ff72a4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-04T06:47:58Z","title_canon_sha256":"f82e8407a9d9c13603e48e95a93a6aad9f964dd76551332776898f1e5c8de739"},"schema_version":"1.0","source":{"id":"2310.03051","kind":"arxiv","version":1}},"canonical_sha256":"9fe2c3ea7b91a34c9062d9ede999e98991757854ba58c8e1b879a08145483ded","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9fe2c3ea7b91a34c9062d9ede999e98991757854ba58c8e1b879a08145483ded","first_computed_at":"2026-07-05T06:57:28.270469Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:57:28.270469Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dYJCzY7rcQTAMvaigLqcKtUd2jQTU6eYMxTP6VA+94AiZk4zALQw5aQiiWWfqzuKejzos/ByPw1oQZcrros5AA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:57:28.270959Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.03051","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b7fcb3214d832b8454213b14b0c9087e94b74c1ba39a9c1411ae7596a3499a9c","sha256:52595757ababf388355db3ddddd4d16fe75b265db400e8fb31e90574285e773a"],"state_sha256":"a022439f48df588ca7b08dda3eb602ccbd35872a54bc4d83f793024d316a38c8"}