{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QTP4WLXP5DPZUTUUCTYUOFVWXS","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":"b2279eb1511fa93da8089a7740e9e387e9b4c5e19e018e0f0b32220b960873b1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-06-11T06:48:57Z","title_canon_sha256":"56ff6802a4b4f66a1f57dd3bb610e796839fb5b18e87429dc6287904fec4544c"},"schema_version":"1.0","source":{"id":"2506.09443","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09443","created_at":"2026-08-07T00:51:05Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09443v3","created_at":"2026-08-07T00:51:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09443","created_at":"2026-08-07T00:51:05Z"},{"alias_kind":"pith_short_12","alias_value":"QTP4WLXP5DPZ","created_at":"2026-08-07T00:51:05Z"},{"alias_kind":"pith_short_16","alias_value":"QTP4WLXP5DPZUTUU","created_at":"2026-08-07T00:51:05Z"},{"alias_kind":"pith_short_8","alias_value":"QTP4WLXP","created_at":"2026-08-07T00:51:05Z"}],"graph_snapshots":[{"event_id":"sha256:ad23fb35eb26464d026ac23581f1a1212bc0eabd26d83abc912a8c7d25c7e047","target":"graph","created_at":"2026-08-07T00:51:05Z","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/2506.09443/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated exceptional capabilities across diverse tasks, driving the development and widespread adoption of LLM-as-a-Judge systems for automated evaluation, including red teaming and benchmarking. However, these systems are susceptible to adversarial attacks that can manipulate evaluation outcomes, raising critical concerns about their robustness and trustworthiness. Existing evaluation methods for LLM-based judges are often fragmented and lack a unified framework for comprehensive robustness assessment. Furthermore, the impact of prompt template design and","authors_text":"Chen Chen, Chuokun Xu, Jiaying Wang, Jirui Zhang, Jun Wang, Kwok-Yan Lam, Shouling Ji, Songze Li, Xueluan Gong","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-06-11T06:48:57Z","title":"LLMs Cannot Reliably Judge (Yet?): A Comprehensive Assessment on the Robustness of LLM-as-a-Judge"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09443","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":"sha256:0512a5ec2845fe6959b88b985f55c243d75ff6a2ca9232de546659388c254bb6","target":"record","created_at":"2026-08-07T00:51:05Z","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":"b2279eb1511fa93da8089a7740e9e387e9b4c5e19e018e0f0b32220b960873b1","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-06-11T06:48:57Z","title_canon_sha256":"56ff6802a4b4f66a1f57dd3bb610e796839fb5b18e87429dc6287904fec4544c"},"schema_version":"1.0","source":{"id":"2506.09443","kind":"arxiv","version":3}},"canonical_sha256":"84dfcb2eefe8df9a4e9414f14716b6bcabb4b6fb7ed294aac94e6c6f4c5155eb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"84dfcb2eefe8df9a4e9414f14716b6bcabb4b6fb7ed294aac94e6c6f4c5155eb","first_computed_at":"2026-08-07T00:51:05.938652Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-07T00:51:05.938652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SDucXdi6MZWzch/v0Bo5fuw25oS1bHrNdlJCn6gVkgL19+MokZmoDwZgVde6rCql7mMWX/lHRR4UhYGTec+fCg==","signature_status":"signed_v1","signed_at":"2026-08-07T00:51:05.940218Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.09443","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0512a5ec2845fe6959b88b985f55c243d75ff6a2ca9232de546659388c254bb6","sha256:ad23fb35eb26464d026ac23581f1a1212bc0eabd26d83abc912a8c7d25c7e047"],"state_sha256":"be29e60f15b4a185e369a67f02dd9fb2cb513d21053d6bbf506c59dcf364b02b"}