{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RRWWA7SR75IM47GCA4PVSPZ44K","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":"34c0eae91b2e2901894adc8a458477028d9073766ca3b9621e4cdb667d748e49","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T06:42:12Z","title_canon_sha256":"cb3e0cc0c583739f6a069ba987c64d32e56741bc1c1628e448382ca16d6f38a0"},"schema_version":"1.0","source":{"id":"2606.09092","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.09092","created_at":"2026-06-09T02:07:58Z"},{"alias_kind":"arxiv_version","alias_value":"2606.09092v1","created_at":"2026-06-09T02:07:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09092","created_at":"2026-06-09T02:07:58Z"},{"alias_kind":"pith_short_12","alias_value":"RRWWA7SR75IM","created_at":"2026-06-09T02:07:58Z"},{"alias_kind":"pith_short_16","alias_value":"RRWWA7SR75IM47GC","created_at":"2026-06-09T02:07:58Z"},{"alias_kind":"pith_short_8","alias_value":"RRWWA7SR","created_at":"2026-06-09T02:07:58Z"}],"graph_snapshots":[{"event_id":"sha256:0a183329cc584190197fb7bb9af4857e64136bae8c77b5846ad695ca3fa7a56f","target":"graph","created_at":"2026-06-09T02:07:58Z","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/2606.09092/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Theory of Mind (ToM) is a must-acquire skill for modern foundation model systems to operate effectively and safely in the real world. Recent works have explored honing ToM via post-training; however, we show that such progress is confounded by a pervasive \"shortcut\" issue: tasks can reach up to 99% accuracy by simply exploiting spurious causal correlations, leading to a false sense of ToM. Motivated by this, we first develop a framework to systematically examine ToM datasets for shortcuts and provide guidance for future development. We find that questions reducible to pure state tracking, such","authors_text":"Behzad Dariush, Jike Zhong, Konstantinos Psounis, Ming Li, Shao-Yuan Lo, Wuao Liu, Yuheng Li, Yuxiang Lai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T06:42:12Z","title":"From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09092","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":"sha256:1569bedb1a332a0c668bfe34c2713aadbe12397493c2479055f31c502c55c55e","target":"record","created_at":"2026-06-09T02:07:58Z","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":"34c0eae91b2e2901894adc8a458477028d9073766ca3b9621e4cdb667d748e49","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T06:42:12Z","title_canon_sha256":"cb3e0cc0c583739f6a069ba987c64d32e56741bc1c1628e448382ca16d6f38a0"},"schema_version":"1.0","source":{"id":"2606.09092","kind":"arxiv","version":1}},"canonical_sha256":"8c6d607e51ff50ce7cc2071f593f3ce2b67ccc975cbd12fe76c8ec259def3530","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8c6d607e51ff50ce7cc2071f593f3ce2b67ccc975cbd12fe76c8ec259def3530","first_computed_at":"2026-06-09T02:07:58.532831Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T02:07:58.532831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3cxsR1cnwQGC2araKGQBU1EzEDWG3c4PaolNW6omBR2ncD+a5UdZQ5gU7cJX2sl6svlxm241etDPb/p4qdG5BA==","signature_status":"signed_v1","signed_at":"2026-06-09T02:07:58.533620Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.09092","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1569bedb1a332a0c668bfe34c2713aadbe12397493c2479055f31c502c55c55e","sha256:0a183329cc584190197fb7bb9af4857e64136bae8c77b5846ad695ca3fa7a56f"],"state_sha256":"ba077eb2bc57d14a164980966c2e1e7d3d226be9e541bedaf34bf0d3bb563111"}