{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LEWKZJEKTPEQOWRI5BBWDCASMQ","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":"20f27ae8f2afbebe71ebb6df12dc787ceb15a012a0b2ee5a439734c5046884e8","cross_cats_sorted":["cs.AI","cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-05-16T15:31:08Z","title_canon_sha256":"ab6ec9f749de338bcbcd2451c7a81d2bc86f230688ab480cb184bd8c8dd007e2"},"schema_version":"1.0","source":{"id":"2505.11365","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11365","created_at":"2026-07-05T11:09:47Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11365v4","created_at":"2026-07-05T11:09:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11365","created_at":"2026-07-05T11:09:47Z"},{"alias_kind":"pith_short_12","alias_value":"LEWKZJEKTPEQ","created_at":"2026-07-05T11:09:47Z"},{"alias_kind":"pith_short_16","alias_value":"LEWKZJEKTPEQOWRI","created_at":"2026-07-05T11:09:47Z"},{"alias_kind":"pith_short_8","alias_value":"LEWKZJEK","created_at":"2026-07-05T11:09:47Z"}],"graph_snapshots":[{"event_id":"sha256:5cfe000f5d8a0110113b21fd0eb7ecf4130bb7e314c339877040fd166f1f11c9","target":"graph","created_at":"2026-07-05T11:09:47Z","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/2505.11365/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Ensuring the safety of large language models (LLMs) is critical for responsible deployment, yet existing evaluations often prioritize performance over identifying failure modes. We introduce Phare, a multilingual diagnostic framework to probe and evaluate LLM behavior across three critical dimensions: hallucination and reliability, social biases, and harmful content generation. Our evaluation of 17 state-of-the-art LLMs reveals patterns of systematic vulnerabilities across all safety dimensions, including sycophancy, prompt sensitivity, and stereotype reproduction. By highlighting these specif","authors_text":"Beno\\^it Mal\\'ezieux, Matteo Dora, Pierre Le Jeune, Weixuan Xiao","cross_cats":["cs.AI","cs.CL","cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-05-16T15:31:08Z","title":"Phare: A Safety Probe for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11365","kind":"arxiv","version":4},"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:b8a03cbf35ea33f033e060c631fa2c33008fa9d95dabdcd23a5ae56b002ac337","target":"record","created_at":"2026-07-05T11:09:47Z","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":"20f27ae8f2afbebe71ebb6df12dc787ceb15a012a0b2ee5a439734c5046884e8","cross_cats_sorted":["cs.AI","cs.CL","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-05-16T15:31:08Z","title_canon_sha256":"ab6ec9f749de338bcbcd2451c7a81d2bc86f230688ab480cb184bd8c8dd007e2"},"schema_version":"1.0","source":{"id":"2505.11365","kind":"arxiv","version":4}},"canonical_sha256":"592caca48a9bc9075a28e8436188126420ec913e92da4b7744efad70108a100e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"592caca48a9bc9075a28e8436188126420ec913e92da4b7744efad70108a100e","first_computed_at":"2026-07-05T11:09:47.151713Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:47.151713Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aulAOxJPFjNOEIDS/g1+iQ07czgzwNjF9z4+OgdzDm285cv0A9Qk+88NMi52TXhdcXjxmneSLHHKSvwQMeLZDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:47.152222Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11365","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b8a03cbf35ea33f033e060c631fa2c33008fa9d95dabdcd23a5ae56b002ac337","sha256:5cfe000f5d8a0110113b21fd0eb7ecf4130bb7e314c339877040fd166f1f11c9"],"state_sha256":"dbbdf8659479b36f74c33746904800153c37f255f5297dd5f9b8dbcaecde8ced"}