{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NUKS2PFWSNY54SPWFDJDVHY2UD","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":"cfcc597e564e60daf1746be001d1ad5e5633002b3e5caeef310b24aa53f93517","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-04T19:05:10Z","title_canon_sha256":"1aece65f59d09ed87e4786440f871986dac281b8975bfe73eca8afd18d4b216e"},"schema_version":"1.0","source":{"id":"2406.02721","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.02721","created_at":"2026-07-05T09:19:28Z"},{"alias_kind":"arxiv_version","alias_value":"2406.02721v3","created_at":"2026-07-05T09:19:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.02721","created_at":"2026-07-05T09:19:28Z"},{"alias_kind":"pith_short_12","alias_value":"NUKS2PFWSNY5","created_at":"2026-07-05T09:19:28Z"},{"alias_kind":"pith_short_16","alias_value":"NUKS2PFWSNY54SPW","created_at":"2026-07-05T09:19:28Z"},{"alias_kind":"pith_short_8","alias_value":"NUKS2PFW","created_at":"2026-07-05T09:19:28Z"}],"graph_snapshots":[{"event_id":"sha256:b229380fc5ab28ffca84634a93f2d403a0b2493030f10d4d033154a246b49582","target":"graph","created_at":"2026-07-05T09:19:28Z","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/2406.02721/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We propose SelfControl, an inference-time model control method utilizing gradients to control the behavior of large language models (LLMs) without explicit human annotations. Given a desired behavior expressed in a natural language suffix string concatenated to the input prompt, SelfControl computes gradients of the LLM's self-evaluation of the suffix with respect to its latent representations. The gradients are used to directly control the auto-regressive generation process towards desired behaviors, which eliminates human supervision, achieves precise and transparent control, and offers on-t","authors_text":"Dan Zhang, Difan Zou, Fan Yin, Min Cai, Shichang Zhang, Yisong Yue, Yuchen Zhang, Ziniu Hu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-04T19:05:10Z","title":"Self-Control of LLM Behaviors by Compressing Suffix Gradient into Prefix Controller"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.02721","kind":"arxiv","version":3},"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":"sha256:6fab36b0f3f1daef3a57797dd1d8a0927f992b1f2f01f485aa2d3cabb74e5927","target":"record","created_at":"2026-07-05T09:19:28Z","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":"cfcc597e564e60daf1746be001d1ad5e5633002b3e5caeef310b24aa53f93517","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-04T19:05:10Z","title_canon_sha256":"1aece65f59d09ed87e4786440f871986dac281b8975bfe73eca8afd18d4b216e"},"schema_version":"1.0","source":{"id":"2406.02721","kind":"arxiv","version":3}},"canonical_sha256":"6d152d3cb69371de49f628d23a9f1aa0f4df63c3f37372b19fad1b6266de97e6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d152d3cb69371de49f628d23a9f1aa0f4df63c3f37372b19fad1b6266de97e6","first_computed_at":"2026-07-05T09:19:28.850448Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:19:28.850448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nIXxCPU7x9FW3r+flXtz+Q+Kp9AVHTmwEnXvCuAe2IHZKCZntLMCZUz/EU72s+X2Ns+Vpcdb/634h47O6KXvAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:19:28.850850Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.02721","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6fab36b0f3f1daef3a57797dd1d8a0927f992b1f2f01f485aa2d3cabb74e5927","sha256:b229380fc5ab28ffca84634a93f2d403a0b2493030f10d4d033154a246b49582"],"state_sha256":"4a3fcdb29af9714536f28abde830301dbe07ea331c680d94dc44212076642f4c"}