{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PPJ5HAHNADZHOPPO7YGTAOL222","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":"ea7e0827b3e797bd72c66b74a80f34912c083e65b01cd41f22e96cbbf5a787c7","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T09:20:38Z","title_canon_sha256":"e1788c36fea24cbd0bf06411e5274c29eba0129e15b98517196cad8a6f134c05"},"schema_version":"1.0","source":{"id":"2607.21090","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21090","created_at":"2026-07-24T01:23:51Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21090v1","created_at":"2026-07-24T01:23:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21090","created_at":"2026-07-24T01:23:51Z"},{"alias_kind":"pith_short_12","alias_value":"PPJ5HAHNADZH","created_at":"2026-07-24T01:23:51Z"},{"alias_kind":"pith_short_16","alias_value":"PPJ5HAHNADZHOPPO","created_at":"2026-07-24T01:23:51Z"},{"alias_kind":"pith_short_8","alias_value":"PPJ5HAHN","created_at":"2026-07-24T01:23:51Z"}],"graph_snapshots":[{"event_id":"sha256:015004f9b7d299e60dd99b6077f8f16498af6f1b5917f4c7e4668aac09717e87","target":"graph","created_at":"2026-07-24T01:23:51Z","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/2607.21090/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose a Reinforcement Learning (RL) method to directly optimize the faithfulness of self-explanations - the extent to which a model's generated reasoning accurately reflects its internal decision-making process. While existing work focuses on evaluating faithfulness or using inference-time prompting frameworks to improve an LLM's self-explanation's tractability, these approaches do not provide a mechanism to directly optimize a model's parameters to generate faithful self-explanations. We bridge this gap by modifying existing faithfulness metrics into an RL training objective. We investig","authors_text":"Mar\\'ia P\\'erez-Ortiz, Noah Y. Siegel, Oana-Maria Camburu, Yeoktatt Cheah","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T09:20:38Z","title":"Training Large Language Models for Self-Explanation Faithfulness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21090","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":"sha256:93d740ac0a7736b33087fefbe79a6fff2a3408fcaa9d39bca98caac2afad19e6","target":"record","created_at":"2026-07-24T01:23:51Z","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":"ea7e0827b3e797bd72c66b74a80f34912c083e65b01cd41f22e96cbbf5a787c7","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T09:20:38Z","title_canon_sha256":"e1788c36fea24cbd0bf06411e5274c29eba0129e15b98517196cad8a6f134c05"},"schema_version":"1.0","source":{"id":"2607.21090","kind":"arxiv","version":1}},"canonical_sha256":"7bd3d380ed00f2773deefe0d30397ad6b45fbb3a75da9c6cb572e58a763272c7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7bd3d380ed00f2773deefe0d30397ad6b45fbb3a75da9c6cb572e58a763272c7","first_computed_at":"2026-07-24T01:23:51.321799Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-24T01:23:51.321799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HiKXg4WDifuQqWIZVI+ym0BxxbStwXI2rEoAnAM5mk2onwN4HEgrF8OxRQTTM2iEmlcocyN+VdQhFjhRUUR3CQ==","signature_status":"signed_v1","signed_at":"2026-07-24T01:23:51.322716Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21090","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:93d740ac0a7736b33087fefbe79a6fff2a3408fcaa9d39bca98caac2afad19e6","sha256:015004f9b7d299e60dd99b6077f8f16498af6f1b5917f4c7e4668aac09717e87"],"state_sha256":"c1d2510e77dc954cba1f989c881d271b8014bcff0de4be8428db43e3301819b9"}