{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FFWBTUFHLNYTYOIZPEJTYCCHDR","short_pith_number":"pith:FFWBTUFH","canonical_record":{"source":{"id":"2506.09450","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T06:55:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"23cbc1d2949f073db4f74a14194a94e6d8a3796f02b37e519b53e02b5254b795","abstract_canon_sha256":"8f465b8381aacdb691ddbba2e21fd33e76175e72c6daef88d36c7faa19cb631a"},"schema_version":"1.0"},"canonical_sha256":"296c19d0a75b713c391979133c08471c799980982b9f9a1e93f54d52ced39ca4","source":{"kind":"arxiv","id":"2506.09450","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09450","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09450v1","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09450","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_12","alias_value":"FFWBTUFHLNYT","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_16","alias_value":"FFWBTUFHLNYTYOIZ","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_8","alias_value":"FFWBTUFH","created_at":"2026-07-05T11:19:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FFWBTUFHLNYTYOIZPEJTYCCHDR","target":"record","payload":{"canonical_record":{"source":{"id":"2506.09450","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T06:55:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"23cbc1d2949f073db4f74a14194a94e6d8a3796f02b37e519b53e02b5254b795","abstract_canon_sha256":"8f465b8381aacdb691ddbba2e21fd33e76175e72c6daef88d36c7faa19cb631a"},"schema_version":"1.0"},"canonical_sha256":"296c19d0a75b713c391979133c08471c799980982b9f9a1e93f54d52ced39ca4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:37.652495Z","signature_b64":"cH9+4HP3wEqKvBZ6GP+fPkLjtN3OpGHGXFif/JUkrVnZ1y9gqqH/MxGEZAhQOn3AhPZ2FL45KSnRwO/eEmOIDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"296c19d0a75b713c391979133c08471c799980982b9f9a1e93f54d52ced39ca4","last_reissued_at":"2026-07-05T11:19:37.651890Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:37.651890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.09450","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:19:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FaugQagZ5AHYP8QMXcgvkowRDOgrredFGngYl3wD1rUijZQmzfbaxYwtVwy2c2HCOOGOYNnNj2IFBUrmtB6tBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T23:37:30.915831Z"},"content_sha256":"f3788036e298c8527bd8d4169cf8c1c4f3d4a41d268a6a21d3c5ef46a5595801","schema_version":"1.0","event_id":"sha256:f3788036e298c8527bd8d4169cf8c1c4f3d4a41d268a6a21d3c5ef46a5595801"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FFWBTUFHLNYTYOIZPEJTYCCHDR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"UniToMBench: Integrating Perspective-Taking to Improve Theory of Mind in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chris Kim, Kevin Zhu, Nitin Yarlagadda, Prameshwar Thiyagarajan, Shamant Sai, Soumil Garg, Vaishnavi Parimi, Zhangir Meirbek","submitted_at":"2025-06-11T06:55:40Z","abstract_excerpt":"Theory of Mind (ToM), the ability to understand the mental states of oneself and others, remains a challenging area for large language models (LLMs), which often fail to predict human mental states accurately. In this paper, we introduce UniToMBench, a unified benchmark that integrates the strengths of SimToM and TOMBENCH to systematically improve and assess ToM capabilities in LLMs by integrating multi-interaction task designs and evolving story scenarios. Supported by a custom dataset of over 1,000 hand-written scenarios, UniToMBench combines perspective-taking techniques with diverse evalua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09450","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/2506.09450/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:19:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K/FC1P9KmSdynlaXwhxjVcFdMJ64LiMe1erYuF1Jyx9CIItPycf5O3hQw2SDBfw8iI1qyBluQTxmOQOXGoq2Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T23:37:30.916359Z"},"content_sha256":"17cd20d2ba64d96facec16a36a9ffd5b0ffe99ab0e44115da6ccc349c3486f4a","schema_version":"1.0","event_id":"sha256:17cd20d2ba64d96facec16a36a9ffd5b0ffe99ab0e44115da6ccc349c3486f4a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/bundle.json","state_url":"https://pith.science/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/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":"2026-08-08T23:37:30Z","links":{"resolver":"https://pith.science/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR","bundle":"https://pith.science/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/bundle.json","state":"https://pith.science/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FFWBTUFHLNYTYOIZPEJTYCCHDR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FFWBTUFHLNYTYOIZPEJTYCCHDR","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":"8f465b8381aacdb691ddbba2e21fd33e76175e72c6daef88d36c7faa19cb631a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T06:55:40Z","title_canon_sha256":"23cbc1d2949f073db4f74a14194a94e6d8a3796f02b37e519b53e02b5254b795"},"schema_version":"1.0","source":{"id":"2506.09450","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09450","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09450v1","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09450","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_12","alias_value":"FFWBTUFHLNYT","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_16","alias_value":"FFWBTUFHLNYTYOIZ","created_at":"2026-07-05T11:19:37Z"},{"alias_kind":"pith_short_8","alias_value":"FFWBTUFH","created_at":"2026-07-05T11:19:37Z"}],"graph_snapshots":[{"event_id":"sha256:17cd20d2ba64d96facec16a36a9ffd5b0ffe99ab0e44115da6ccc349c3486f4a","target":"graph","created_at":"2026-07-05T11:19:37Z","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/2506.09450/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Theory of Mind (ToM), the ability to understand the mental states of oneself and others, remains a challenging area for large language models (LLMs), which often fail to predict human mental states accurately. In this paper, we introduce UniToMBench, a unified benchmark that integrates the strengths of SimToM and TOMBENCH to systematically improve and assess ToM capabilities in LLMs by integrating multi-interaction task designs and evolving story scenarios. Supported by a custom dataset of over 1,000 hand-written scenarios, UniToMBench combines perspective-taking techniques with diverse evalua","authors_text":"Chris Kim, Kevin Zhu, Nitin Yarlagadda, Prameshwar Thiyagarajan, Shamant Sai, Soumil Garg, Vaishnavi Parimi, Zhangir Meirbek","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T06:55:40Z","title":"UniToMBench: Integrating Perspective-Taking to Improve Theory of Mind in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09450","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:f3788036e298c8527bd8d4169cf8c1c4f3d4a41d268a6a21d3c5ef46a5595801","target":"record","created_at":"2026-07-05T11:19:37Z","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":"8f465b8381aacdb691ddbba2e21fd33e76175e72c6daef88d36c7faa19cb631a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T06:55:40Z","title_canon_sha256":"23cbc1d2949f073db4f74a14194a94e6d8a3796f02b37e519b53e02b5254b795"},"schema_version":"1.0","source":{"id":"2506.09450","kind":"arxiv","version":1}},"canonical_sha256":"296c19d0a75b713c391979133c08471c799980982b9f9a1e93f54d52ced39ca4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"296c19d0a75b713c391979133c08471c799980982b9f9a1e93f54d52ced39ca4","first_computed_at":"2026-07-05T11:19:37.651890Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:37.651890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cH9+4HP3wEqKvBZ6GP+fPkLjtN3OpGHGXFif/JUkrVnZ1y9gqqH/MxGEZAhQOn3AhPZ2FL45KSnRwO/eEmOIDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:37.652495Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.09450","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f3788036e298c8527bd8d4169cf8c1c4f3d4a41d268a6a21d3c5ef46a5595801","sha256:17cd20d2ba64d96facec16a36a9ffd5b0ffe99ab0e44115da6ccc349c3486f4a"],"state_sha256":"8e205811a185004fff06b3b44f298f8315846a25ad062e2b2d0056e734ae8296"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ICJWJchF1RQpoTr5hflBRbnrBrm3WY5GSkJfJzQn9FZnnG0bJxEguJTmYAA1Ph7xiQ7AQ39LkQcOJY/P1g4xDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T23:37:30.921043Z","bundle_sha256":"e094124ed0d079cb2a766aad9956d26384a785f46fb20bd8383aba7029466ba9"}}