{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YXZ4JDPRXCLMU4MXAKZDBTIOLE","short_pith_number":"pith:YXZ4JDPR","canonical_record":{"source":{"id":"2508.06296","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-08T13:15:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"28bf7510dd445e4749cdc8f6ec7777cd42cdd5b14c78932f7978f579f38629cd","abstract_canon_sha256":"90cbf67f8b608769929aa15a7569f80119508d519f8611ef1958ab024a28fa95"},"schema_version":"1.0"},"canonical_sha256":"c5f3c48df1b896ca719702b230cd0e593a3bb436da19ba9382de86bc3765eef9","source":{"kind":"arxiv","id":"2508.06296","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06296","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06296v2","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06296","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_12","alias_value":"YXZ4JDPRXCLM","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_16","alias_value":"YXZ4JDPRXCLMU4MX","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_8","alias_value":"YXZ4JDPR","created_at":"2026-07-05T11:53:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YXZ4JDPRXCLMU4MXAKZDBTIOLE","target":"record","payload":{"canonical_record":{"source":{"id":"2508.06296","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-08T13:15:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"28bf7510dd445e4749cdc8f6ec7777cd42cdd5b14c78932f7978f579f38629cd","abstract_canon_sha256":"90cbf67f8b608769929aa15a7569f80119508d519f8611ef1958ab024a28fa95"},"schema_version":"1.0"},"canonical_sha256":"c5f3c48df1b896ca719702b230cd0e593a3bb436da19ba9382de86bc3765eef9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:07.194678Z","signature_b64":"zUygfwFfiQ5zfGtEvMcr0wLGn50cL/s0182P1mtIBuT3nNC+TGt0BtZGE5DatHKNnMIx29fysFXnyISLpOIXAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5f3c48df1b896ca719702b230cd0e593a3bb436da19ba9382de86bc3765eef9","last_reissued_at":"2026-07-05T11:53:07.194175Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:07.194175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.06296","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:53:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AZPv6s0zHTkdDFt3oWPSeaWd8SZGPZL1jXuIJMxDyTV0qZTvCuIE6fhzEekhQAfmL5ZP0mW/hBTtK3qTeiHZAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T16:01:54.280704Z"},"content_sha256":"37121df7f0ea9b60bd41efc8595fba424e40835c5cea00d4ac1b0a1c16616eac","schema_version":"1.0","event_id":"sha256:37121df7f0ea9b60bd41efc8595fba424e40835c5cea00d4ac1b0a1c16616eac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YXZ4JDPRXCLMU4MXAKZDBTIOLE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLM Robustness Leaderboard v1 --Technical report","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Nicolas Miailhe, Pierre Peign\\'e - Lefebvre, Quentin Feuillade-Montixi, Tom David","submitted_at":"2025-08-08T13:15:40Z","abstract_excerpt":"This technical report accompanies the LLM robustness leaderboard published by PRISM Eval for the Paris AI Action Summit. We introduce PRISM Eval Behavior Elicitation Tool (BET), an AI system performing automated red-teaming through Dynamic Adversarial Optimization that achieves 100% Attack Success Rate (ASR) against 37 of 41 state-of-the-art LLMs. Beyond binary success metrics, we propose a fine-grained robustness metric estimating the average number of attempts required to elicit harmful behaviors, revealing that attack difficulty varies by over 300-fold across models despite universal vulner"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06296","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.06296/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:53:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pUh8Lj/wPWi13LXWw2QeiCMDgz8XK0Kg6Pk/JS5uE9cE5Svq704IZrrTbZnhEQlhXS9ykLfkDqoAUAx3j8oSAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-08T16:01:54.281089Z"},"content_sha256":"464565a8963e6d6f240e35a44ec02e43c33322f6b2be6e4bd5bce829817b070b","schema_version":"1.0","event_id":"sha256:464565a8963e6d6f240e35a44ec02e43c33322f6b2be6e4bd5bce829817b070b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/bundle.json","state_url":"https://pith.science/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-08T16:01:54Z","links":{"resolver":"https://pith.science/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE","bundle":"https://pith.science/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/bundle.json","state":"https://pith.science/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YXZ4JDPRXCLMU4MXAKZDBTIOLE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YXZ4JDPRXCLMU4MXAKZDBTIOLE","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":"90cbf67f8b608769929aa15a7569f80119508d519f8611ef1958ab024a28fa95","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-08T13:15:40Z","title_canon_sha256":"28bf7510dd445e4749cdc8f6ec7777cd42cdd5b14c78932f7978f579f38629cd"},"schema_version":"1.0","source":{"id":"2508.06296","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06296","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06296v2","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06296","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_12","alias_value":"YXZ4JDPRXCLM","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_16","alias_value":"YXZ4JDPRXCLMU4MX","created_at":"2026-07-05T11:53:07Z"},{"alias_kind":"pith_short_8","alias_value":"YXZ4JDPR","created_at":"2026-07-05T11:53:07Z"}],"graph_snapshots":[{"event_id":"sha256:464565a8963e6d6f240e35a44ec02e43c33322f6b2be6e4bd5bce829817b070b","target":"graph","created_at":"2026-07-05T11:53:07Z","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/2508.06296/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This technical report accompanies the LLM robustness leaderboard published by PRISM Eval for the Paris AI Action Summit. We introduce PRISM Eval Behavior Elicitation Tool (BET), an AI system performing automated red-teaming through Dynamic Adversarial Optimization that achieves 100% Attack Success Rate (ASR) against 37 of 41 state-of-the-art LLMs. Beyond binary success metrics, we propose a fine-grained robustness metric estimating the average number of attempts required to elicit harmful behaviors, revealing that attack difficulty varies by over 300-fold across models despite universal vulner","authors_text":"Nicolas Miailhe, Pierre Peign\\'e - Lefebvre, Quentin Feuillade-Montixi, Tom David","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-08T13:15:40Z","title":"LLM Robustness Leaderboard v1 --Technical report"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06296","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":"sha256:37121df7f0ea9b60bd41efc8595fba424e40835c5cea00d4ac1b0a1c16616eac","target":"record","created_at":"2026-07-05T11:53:07Z","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":"90cbf67f8b608769929aa15a7569f80119508d519f8611ef1958ab024a28fa95","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-08T13:15:40Z","title_canon_sha256":"28bf7510dd445e4749cdc8f6ec7777cd42cdd5b14c78932f7978f579f38629cd"},"schema_version":"1.0","source":{"id":"2508.06296","kind":"arxiv","version":2}},"canonical_sha256":"c5f3c48df1b896ca719702b230cd0e593a3bb436da19ba9382de86bc3765eef9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c5f3c48df1b896ca719702b230cd0e593a3bb436da19ba9382de86bc3765eef9","first_computed_at":"2026-07-05T11:53:07.194175Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:53:07.194175Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zUygfwFfiQ5zfGtEvMcr0wLGn50cL/s0182P1mtIBuT3nNC+TGt0BtZGE5DatHKNnMIx29fysFXnyISLpOIXAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:53:07.194678Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.06296","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:37121df7f0ea9b60bd41efc8595fba424e40835c5cea00d4ac1b0a1c16616eac","sha256:464565a8963e6d6f240e35a44ec02e43c33322f6b2be6e4bd5bce829817b070b"],"state_sha256":"6e65cd28aa74d94b8c2ef0164d1e8b687da4c2c94d4ff0afc7b206dad651bb49"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LegcXV1JSIuEwW9335dseYSTPdO3ZnCem/bsdCxhiTP4chj5JGRMqN2/5/gmvpe6QcbmFaDWY1Z43mvXl5oXBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-08T16:01:54.283331Z","bundle_sha256":"c2aaf0bc8fe0f056023d4147312473c7c22eaf122b5c763e6e492c762f9a758a"}}