{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TQLEP4SMFW35LYST3FQ55SVWDC","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":"8bcffb2121093e1e8f8665448b896c5f380db15a65712220334d20bc50615343","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-30T13:38:06Z","title_canon_sha256":"84841c0c41c86532509dfbe2a1129583e4df4d1198c3a74a1ba57f960a3aa2a2"},"schema_version":"1.0","source":{"id":"2606.00726","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.00726","created_at":"2026-06-02T01:04:03Z"},{"alias_kind":"arxiv_version","alias_value":"2606.00726v1","created_at":"2026-06-02T01:04:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00726","created_at":"2026-06-02T01:04:03Z"},{"alias_kind":"pith_short_12","alias_value":"TQLEP4SMFW35","created_at":"2026-06-02T01:04:03Z"},{"alias_kind":"pith_short_16","alias_value":"TQLEP4SMFW35LYST","created_at":"2026-06-02T01:04:03Z"},{"alias_kind":"pith_short_8","alias_value":"TQLEP4SM","created_at":"2026-06-02T01:04:03Z"}],"graph_snapshots":[{"event_id":"sha256:a46fa53af0ab9736ac0bcca64acaf12244a8e123b20b661caa69e98e6cc60a0c","target":"graph","created_at":"2026-06-02T01:04:03Z","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/2606.00726/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Strong reasoning depends not only on model knowledge but also on how effectively cognitive behaviors are deployed during generation. Existing methods often rely on explicit behavior-level control, making them insufficiently adaptive when failures and required corrections vary across reasoning states, tasks, and models. To this end, we propose Latent Reward Steering (LRS), an adaptive inference-time framework that promotes cognitive behaviors by optimizing the sparse-autoencoder (SAE) latent states that implicitly carry them. Rather than relying on predefined cognitive behaviors or steering dir","authors_text":"Can Jin, Chenxi Huang, Dexu Yu, Dimitris N. Metaxas, Guanyu Zhu, Hongwu Peng, Jiakang Li, Ronghao Chen, Xuanqi Lan, Yang Zhou, Youhua Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-30T13:38:06Z","title":"Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00726","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":"sha256:0223c70077e281dad841cf0d0eda4b1b59d14dd803225abb3346c736166a12db","target":"record","created_at":"2026-06-02T01:04:03Z","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":"8bcffb2121093e1e8f8665448b896c5f380db15a65712220334d20bc50615343","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-30T13:38:06Z","title_canon_sha256":"84841c0c41c86532509dfbe2a1129583e4df4d1198c3a74a1ba57f960a3aa2a2"},"schema_version":"1.0","source":{"id":"2606.00726","kind":"arxiv","version":1}},"canonical_sha256":"9c1647f24c2db7d5e253d961decab6188bccf037eda8360a1962ed63b8954420","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c1647f24c2db7d5e253d961decab6188bccf037eda8360a1962ed63b8954420","first_computed_at":"2026-06-02T01:04:03.958443Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:04:03.958443Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"e2HwDFV+KZ7rv4S6IhKrxIxCsJpZLPjwsj+xmaed0i7WuJ7TbWkahjWk8PRDFD4l//yVyto0XjSV1cE5of/HBA==","signature_status":"signed_v1","signed_at":"2026-06-02T01:04:03.958847Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.00726","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0223c70077e281dad841cf0d0eda4b1b59d14dd803225abb3346c736166a12db","sha256:a46fa53af0ab9736ac0bcca64acaf12244a8e123b20b661caa69e98e6cc60a0c"],"state_sha256":"687c0c4ea2b97c3d914d36c0a2e536c31cc2b42d9253e325d5faaa6e05267748"}