{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:3JMZEFV2NOUKAK6VBHRJZ6ES3L","short_pith_number":"pith:3JMZEFV2","canonical_record":{"source":{"id":"2507.15788","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T16:47:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"52898d728d827f9a36b6ceb185fc80b247dcb69f87be3de6a3a661ee6eec8c77","abstract_canon_sha256":"bd25b121fd31b68623bfeb234644fe75b6579c3809eba0bdae9254622398776f"},"schema_version":"1.0"},"canonical_sha256":"da599216ba6ba8a02bd509e29cf892daf685521d996d33df5521a0cce9e0b6af","source":{"kind":"arxiv","id":"2507.15788","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.15788","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"arxiv_version","alias_value":"2507.15788v1","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15788","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_12","alias_value":"3JMZEFV2NOUK","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_16","alias_value":"3JMZEFV2NOUKAK6V","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_8","alias_value":"3JMZEFV2","created_at":"2026-07-05T11:40:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:3JMZEFV2NOUKAK6VBHRJZ6ES3L","target":"record","payload":{"canonical_record":{"source":{"id":"2507.15788","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T16:47:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"52898d728d827f9a36b6ceb185fc80b247dcb69f87be3de6a3a661ee6eec8c77","abstract_canon_sha256":"bd25b121fd31b68623bfeb234644fe75b6579c3809eba0bdae9254622398776f"},"schema_version":"1.0"},"canonical_sha256":"da599216ba6ba8a02bd509e29cf892daf685521d996d33df5521a0cce9e0b6af","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:40:44.280263Z","signature_b64":"9yr3a2//F/SZJ0KFJBXMKWKE4P2Bz1MX0oj/AEZMBSwjnj0Pinc7vqcTDIvTWlL96KlETZH9WWXNS3S2RSx0Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da599216ba6ba8a02bd509e29cf892daf685521d996d33df5521a0cce9e0b6af","last_reissued_at":"2026-07-05T11:40:44.279793Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:40:44.279793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.15788","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":"2026-07-05T11:40:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ojk6l5bBws9e0/rM6aEDEEP8Lf2QZqAg6pWGHLhwJx+dDAgTQmHo5LZgywtaHorV/RuCaJGCFLfi1pClt8SHDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:28:22.870069Z"},"content_sha256":"63cc9538664922e23ea001ea86d33374a0b70e5325531a2b77424360eb1d8e35","schema_version":"1.0","event_id":"sha256:63cc9538664922e23ea001ea86d33374a0b70e5325531a2b77424360eb1d8e35"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:3JMZEFV2NOUKAK6VBHRJZ6ES3L","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Small LLMs Do Not Learn a Generalizable Theory of Mind via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Hanan Salam, Sneheel Sarangi","submitted_at":"2025-07-21T16:47:59Z","abstract_excerpt":"Recent advancements in large language models (LLMs) have demonstrated emergent capabilities in complex reasoning, largely spurred by rule-based Reinforcement Learning (RL) techniques applied during the post-training. This has raised the question of whether similar methods can instill more nuanced, human-like social intelligence, such as a Theory of Mind (ToM), in LLMs. This paper investigates whether small-scale LLMs can acquire a robust and generalizable ToM capability through RL with verifiable rewards (RLVR). We conduct a systematic evaluation by training models on various combinations of p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15788","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/2507.15788/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":"2026-07-05T11:40:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GR6DIuWxNz70uDFXQFZt/jwJtCCVmt3HHX+pFcU8w0AClspUBc1gP4v2pl9UXXuoYeq2xcPJtumubZcqseNkCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:28:22.870986Z"},"content_sha256":"4c1c23bbbfb51be92ae0480cf1b8e98cc6b3360dfd37a15a265702e6850a781a","schema_version":"1.0","event_id":"sha256:4c1c23bbbfb51be92ae0480cf1b8e98cc6b3360dfd37a15a265702e6850a781a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/bundle.json","state_url":"https://pith.science/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/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":"2026-08-07T18:28:22Z","links":{"resolver":"https://pith.science/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L","bundle":"https://pith.science/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/bundle.json","state":"https://pith.science/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3JMZEFV2NOUKAK6VBHRJZ6ES3L/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3JMZEFV2NOUKAK6VBHRJZ6ES3L","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":"bd25b121fd31b68623bfeb234644fe75b6579c3809eba0bdae9254622398776f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T16:47:59Z","title_canon_sha256":"52898d728d827f9a36b6ceb185fc80b247dcb69f87be3de6a3a661ee6eec8c77"},"schema_version":"1.0","source":{"id":"2507.15788","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.15788","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"arxiv_version","alias_value":"2507.15788v1","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.15788","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_12","alias_value":"3JMZEFV2NOUK","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_16","alias_value":"3JMZEFV2NOUKAK6V","created_at":"2026-07-05T11:40:44Z"},{"alias_kind":"pith_short_8","alias_value":"3JMZEFV2","created_at":"2026-07-05T11:40:44Z"}],"graph_snapshots":[{"event_id":"sha256:4c1c23bbbfb51be92ae0480cf1b8e98cc6b3360dfd37a15a265702e6850a781a","target":"graph","created_at":"2026-07-05T11:40:44Z","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/2507.15788/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in large language models (LLMs) have demonstrated emergent capabilities in complex reasoning, largely spurred by rule-based Reinforcement Learning (RL) techniques applied during the post-training. This has raised the question of whether similar methods can instill more nuanced, human-like social intelligence, such as a Theory of Mind (ToM), in LLMs. This paper investigates whether small-scale LLMs can acquire a robust and generalizable ToM capability through RL with verifiable rewards (RLVR). We conduct a systematic evaluation by training models on various combinations of p","authors_text":"Hanan Salam, Sneheel Sarangi","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T16:47:59Z","title":"Small LLMs Do Not Learn a Generalizable Theory of Mind via Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.15788","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":"sha256:63cc9538664922e23ea001ea86d33374a0b70e5325531a2b77424360eb1d8e35","target":"record","created_at":"2026-07-05T11:40:44Z","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":"bd25b121fd31b68623bfeb234644fe75b6579c3809eba0bdae9254622398776f","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-21T16:47:59Z","title_canon_sha256":"52898d728d827f9a36b6ceb185fc80b247dcb69f87be3de6a3a661ee6eec8c77"},"schema_version":"1.0","source":{"id":"2507.15788","kind":"arxiv","version":1}},"canonical_sha256":"da599216ba6ba8a02bd509e29cf892daf685521d996d33df5521a0cce9e0b6af","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da599216ba6ba8a02bd509e29cf892daf685521d996d33df5521a0cce9e0b6af","first_computed_at":"2026-07-05T11:40:44.279793Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:40:44.279793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9yr3a2//F/SZJ0KFJBXMKWKE4P2Bz1MX0oj/AEZMBSwjnj0Pinc7vqcTDIvTWlL96KlETZH9WWXNS3S2RSx0Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T11:40:44.280263Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.15788","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:63cc9538664922e23ea001ea86d33374a0b70e5325531a2b77424360eb1d8e35","sha256:4c1c23bbbfb51be92ae0480cf1b8e98cc6b3360dfd37a15a265702e6850a781a"],"state_sha256":"144984f77b49faa902ee6d90d575426eb03f1fc86f8bdee17544ba9f70365366"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q4K1wc2id/QD76u65thPIGQM/n5uCsQlKS5gjJN912uiFcT2EozR605o4+H7wT9T9uM4sSy7p+SSOpJLQAGYCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T18:28:22.875365Z","bundle_sha256":"a2d6f9d6c8d818e18c0fc657d63eeecf4bf493ab3033379e783b16b9abf5c4da"}}