{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PGFZW4RPFHXBPN6JMCKQLGLASW","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":"196e0579dcd3ffe8c3b495e0f0653d56a2f5c4afa1312aa3bfdbfe37338f843b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-02T05:10:29Z","title_canon_sha256":"da16ce701eb52f1032e30af4dfcab263ad30765c8acffd5672f4030ec3d9b15d"},"schema_version":"1.0","source":{"id":"2507.01368","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.01368","created_at":"2026-07-05T11:30:47Z"},{"alias_kind":"arxiv_version","alias_value":"2507.01368v1","created_at":"2026-07-05T11:30:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.01368","created_at":"2026-07-05T11:30:47Z"},{"alias_kind":"pith_short_12","alias_value":"PGFZW4RPFHXB","created_at":"2026-07-05T11:30:47Z"},{"alias_kind":"pith_short_16","alias_value":"PGFZW4RPFHXBPN6J","created_at":"2026-07-05T11:30:47Z"},{"alias_kind":"pith_short_8","alias_value":"PGFZW4RP","created_at":"2026-07-05T11:30:47Z"}],"graph_snapshots":[{"event_id":"sha256:54a794b68be87df253b5deb3fc59a791ee213a59e3a52635bb302419a2ecf91f","target":"graph","created_at":"2026-07-05T11:30:47Z","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/2507.01368/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Aligning Large Language Models (LLMs) and Large Multimodal Models (LMMs) to human preferences is a central challenge in improving the quality of the models' generative outputs for real-world applications. A common approach is to use reward modeling to encode preferences, enabling alignment via post-training using reinforcement learning. However, traditional reward modeling is not easily adaptable to new preferences because it requires a separate reward model, commonly trained on large preference datasets. To address this, we introduce Activation Reward Models (Activation RMs) -- a novel few-sh","authors_text":"Assaf Arbelle, Brandon Huang, Chancharik Mitra, Deva Ramanan, Gautam Rajendrakumar Gare, Leonid Karlinsky, Roei Herzig, Rogerio Feris, Tianning Chai, Trevor Darrell, Zhiqiu Lin","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-02T05:10:29Z","title":"Activation Reward Models for Few-Shot Model Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.01368","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":"sha256:67739317bd9a968a2aeee68fdda6ade8fa846db46ba7994263f50f38a8a120b9","target":"record","created_at":"2026-07-05T11:30:47Z","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":"196e0579dcd3ffe8c3b495e0f0653d56a2f5c4afa1312aa3bfdbfe37338f843b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-02T05:10:29Z","title_canon_sha256":"da16ce701eb52f1032e30af4dfcab263ad30765c8acffd5672f4030ec3d9b15d"},"schema_version":"1.0","source":{"id":"2507.01368","kind":"arxiv","version":1}},"canonical_sha256":"798b9b722f29ee17b7c9609505996095af9989fc13639dc18d0e75a27808b8c6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"798b9b722f29ee17b7c9609505996095af9989fc13639dc18d0e75a27808b8c6","first_computed_at":"2026-07-05T11:30:47.591926Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:47.591926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MAPQlGz7kS3yh0YyQ2EVZCYbh3F9rONc7xef3LvCFmZwCA6fw2Uj553LUc+tMZ6KOAvbBXDIGjq6bZjZyOjdCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:47.592435Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.01368","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67739317bd9a968a2aeee68fdda6ade8fa846db46ba7994263f50f38a8a120b9","sha256:54a794b68be87df253b5deb3fc59a791ee213a59e3a52635bb302419a2ecf91f"],"state_sha256":"0fbc02f8a5c42bbd820131d3aaf117e4060dae96dabb36ba5cf650625e9358db"}