{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WI67VKJHARZ26PBW5TKG62MVCG","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":"2b337e48ffb43864b346fd5d29457429612f02d6a93dafdcdb2362170bf49dbf","cross_cats_sorted":["cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-19T09:22:11Z","title_canon_sha256":"f45a4de46355cc3ae2cfc8defeb32085472bf5aba97429012b810a8a900cd443"},"schema_version":"1.0","source":{"id":"2512.17375","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.17375","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"arxiv_version","alias_value":"2512.17375v2","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.17375","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_12","alias_value":"WI67VKJHARZ2","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_16","alias_value":"WI67VKJHARZ26PBW","created_at":"2026-05-28T01:04:07Z"},{"alias_kind":"pith_short_8","alias_value":"WI67VKJH","created_at":"2026-05-28T01:04:07Z"}],"graph_snapshots":[{"event_id":"sha256:da94f8e8e7ee4f188805acccb8e589495833d81cff6dfb8f7d0016e55c321987","target":"graph","created_at":"2026-05-28T01:04:07Z","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/2512.17375/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-as-a-Judge systems supply the reward signal in modern RLHF and RLVR pipelines, but their binary verdict reduces to a single linear readout F_gap on one hidden state. We show this readout is shallow enough that short, low-perplexity tokens flip the verdict from \"No\" to \"Yes\". These tokens are sampled from the judge's own next-token distribution at the response position, with no manual seed set and no gradient-based optimization. Our procedure, AdvJudge-Zero, reaches $>$90% ensemble false-positive rate on 22 of 24 (model, dataset) cells across six Qwen, Llama, and Gemma judges, versus 54-72%","authors_text":"Hongliang Liu, Tung-Ling Li, Yuhao Wu","cross_cats":["cs.CL","cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-19T09:22:11Z","title":"AdvJudge-Zero: Binary Decision Flips in LLM-as-a-Judge via Adversarial Control Tokens"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.17375","kind":"arxiv","version":2},"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":"sha256:48fc2b30d722bdc15cc27553ffca9d7565b7d68ed3f65aabcf9ca77a035453c9","target":"record","created_at":"2026-05-28T01:04:07Z","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":"2b337e48ffb43864b346fd5d29457429612f02d6a93dafdcdb2362170bf49dbf","cross_cats_sorted":["cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-19T09:22:11Z","title_canon_sha256":"f45a4de46355cc3ae2cfc8defeb32085472bf5aba97429012b810a8a900cd443"},"schema_version":"1.0","source":{"id":"2512.17375","kind":"arxiv","version":2}},"canonical_sha256":"b23dfaa9270473af3c36ecd46f69951182e38987fbb0a5842ec4226751008acb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b23dfaa9270473af3c36ecd46f69951182e38987fbb0a5842ec4226751008acb","first_computed_at":"2026-05-28T01:04:07.659226Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:07.659226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yoqI+FwMvOBzhnCBlsgPF0LlmKdA9OP9vXtU+IXfZpShnX42m3yxmtg+1r6ymhYUg5uyrI/U3S5bj6MaFGrQAg==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:07.659915Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.17375","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:48fc2b30d722bdc15cc27553ffca9d7565b7d68ed3f65aabcf9ca77a035453c9","sha256:da94f8e8e7ee4f188805acccb8e589495833d81cff6dfb8f7d0016e55c321987"],"state_sha256":"2fe6253010f8f5079e802598a0126e40aed08b7cc599c0306f051939651f8f3f"}