{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SWOHONJNQSBF32OXBWGIQ3RYDD","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":"6657b2240d64785486a42e2bc9061e1e8affe1311a4979b944af489e8f4a1e74","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-25T20:45:08Z","title_canon_sha256":"e30486f4115a85a39e755503deb9c2ed58440eb68ac34e22a7f538f5deddd4de"},"schema_version":"1.0","source":{"id":"2605.26322","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.26322","created_at":"2026-05-27T01:05:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.26322v1","created_at":"2026-05-27T01:05:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26322","created_at":"2026-05-27T01:05:11Z"},{"alias_kind":"pith_short_12","alias_value":"SWOHONJNQSBF","created_at":"2026-05-27T01:05:11Z"},{"alias_kind":"pith_short_16","alias_value":"SWOHONJNQSBF32OX","created_at":"2026-05-27T01:05:11Z"},{"alias_kind":"pith_short_8","alias_value":"SWOHONJN","created_at":"2026-05-27T01:05:11Z"}],"graph_snapshots":[{"event_id":"sha256:a3c2a2c8fc94db0aa325092608cc63f803c082a7743b45279dc50c18a430aad8","target":"graph","created_at":"2026-05-27T01:05:11Z","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/2605.26322/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 infer others' knowledge, intentions, and emotions, is commonly evaluated in large language models (LLMs) using end-point question answering, where performance is judged solely by the final answer to a social reasoning query. This paradigm obscures whether the model actually constructs the underlying mental-state representations required for robust reasoning, particularly in scenarios involving divergent, evolving, or mistaken beliefs. In order to address this research gap, we introduce OmniToM, a benchmark that directly evaluates these representations by re","authors_text":"Adam Bawatneh, Amrit Singh Bedi, Mubarak Shah, Sagar Sapkota, Santu Karmaker","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-25T20:45:08Z","title":"OmniToM: Benchmarking Theory of Mind in LLMs via Explicit Belief Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26322","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":"sha256:751a8d4cbab19d6e03a4c1f065d041fb54bcacee4ffa06afb8d6793551a46f4f","target":"record","created_at":"2026-05-27T01:05:11Z","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":"6657b2240d64785486a42e2bc9061e1e8affe1311a4979b944af489e8f4a1e74","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-25T20:45:08Z","title_canon_sha256":"e30486f4115a85a39e755503deb9c2ed58440eb68ac34e22a7f538f5deddd4de"},"schema_version":"1.0","source":{"id":"2605.26322","kind":"arxiv","version":1}},"canonical_sha256":"959c77352d84825de9d70d8c886e3818fa9c63ef63c84b24164c11601ec87bff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"959c77352d84825de9d70d8c886e3818fa9c63ef63c84b24164c11601ec87bff","first_computed_at":"2026-05-27T01:05:11.978640Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:05:11.978640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ruEnkYwK8cDWu4ZRnmytOfx8mu18XtBG2xVFWtcQhzNAb6uLCbUJW6jqMbYN80J7+Xx6CceDRpI1Ztoin1oPBw==","signature_status":"signed_v1","signed_at":"2026-05-27T01:05:11.979361Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.26322","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:751a8d4cbab19d6e03a4c1f065d041fb54bcacee4ffa06afb8d6793551a46f4f","sha256:a3c2a2c8fc94db0aa325092608cc63f803c082a7743b45279dc50c18a430aad8"],"state_sha256":"c84169d5b5312d71fb349dcb720c52b848c38df7ca1dd07075791e16524a0de3"}