{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:45A7QZU2T2CDNRSFS7RTYCJ6WD","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":"9ce9cf618631e187a3c35e2985c1bddb25280027c02a633b2a72799a012df936","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-31T18:05:25Z","title_canon_sha256":"074dd57fed2edffd5c041726b790c54c79e4749028f7a64bb1940df0ed1c65d7"},"schema_version":"1.0","source":{"id":"2608.00180","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00180","created_at":"2026-08-04T00:33:14Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00180v1","created_at":"2026-08-04T00:33:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00180","created_at":"2026-08-04T00:33:14Z"},{"alias_kind":"pith_short_12","alias_value":"45A7QZU2T2CD","created_at":"2026-08-04T00:33:14Z"},{"alias_kind":"pith_short_16","alias_value":"45A7QZU2T2CDNRSF","created_at":"2026-08-04T00:33:14Z"},{"alias_kind":"pith_short_8","alias_value":"45A7QZU2","created_at":"2026-08-04T00:33:14Z"}],"graph_snapshots":[{"event_id":"sha256:b133e8f4bedeeaa21c0b0071c14d94738e0ef15b278967dd3a2429040f73a22a","target":"graph","created_at":"2026-08-04T00:33:14Z","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/2608.00180/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Conflicting objectives are general in RL alignment, and training on them data-efficiently is hard. Training a safety guard with RL means optimizing two objectives that conflict: catch real harm, and do not refuse benign prompts. Our finding is that over-refusal improves 22.4% to 12.8%, while under-refusal on adversarial attacks silently worsens 0.27 to 0.33. We present C-Guard, a constitution-grid instrument that generates the RL training data, and C-LIM, a per-cell learnability score that decides each cell's move: prune, densify, amend, expand. C-LIM flags the dead-weight data region before a","authors_text":"Lily Zhang","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-31T18:05:25Z","title":"A Constitution-Grid Instrument for Data-Efficient RL Alignment (C-Guard)"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00180","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":"sha256:6a2b2d71c9a44e0f71b838c224132b612907674643ef64a972835432a4a29e85","target":"record","created_at":"2026-08-04T00:33:14Z","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":"9ce9cf618631e187a3c35e2985c1bddb25280027c02a633b2a72799a012df936","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-31T18:05:25Z","title_canon_sha256":"074dd57fed2edffd5c041726b790c54c79e4749028f7a64bb1940df0ed1c65d7"},"schema_version":"1.0","source":{"id":"2608.00180","kind":"arxiv","version":1}},"canonical_sha256":"e741f8669a9e8436c64597e33c093eb0c4ec3656a96a74f1834d29035f33cc7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e741f8669a9e8436c64597e33c093eb0c4ec3656a96a74f1834d29035f33cc7b","first_computed_at":"2026-08-04T00:33:14.714762Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T00:33:14.714762Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ct0HYbSNvh97HFBAKQi55NskVLU1eUIk1XBDXVwaXitR1GUupHpCvOb3G2m8j8gXhUm+7U0vt8VV+ipwQpW0Aw==","signature_status":"signed_v1","signed_at":"2026-08-04T00:33:14.716098Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.00180","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6a2b2d71c9a44e0f71b838c224132b612907674643ef64a972835432a4a29e85","sha256:b133e8f4bedeeaa21c0b0071c14d94738e0ef15b278967dd3a2429040f73a22a"],"state_sha256":"4455cee43219a0faf393d77b50cbb3c624b592ab08a0a55c96f05b4b51feade3"}