{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QH7YROHDI2KMERXY4PYAQHNO55","short_pith_number":"pith:QH7YROHD","canonical_record":{"source":{"id":"2605.19377","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T05:22:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"791a64563c30c11b64e7f92290ffdd42e1b9676b2f69028b4f8ab1c60858fdec","abstract_canon_sha256":"b6f65e463bca20fe0863be78ebdfc87bf6ea164b79f3dc0b0e861e5c86b2bbd6"},"schema_version":"1.0"},"canonical_sha256":"81ff88b8e34694c246f8e3f0081daeef5fe2efa4d845dcff0e40a84d584490f7","source":{"kind":"arxiv","id":"2605.19377","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19377","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19377v1","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19377","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_12","alias_value":"QH7YROHDI2KM","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_16","alias_value":"QH7YROHDI2KMERXY","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_8","alias_value":"QH7YROHD","created_at":"2026-05-20T01:05:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QH7YROHDI2KMERXY4PYAQHNO55","target":"record","payload":{"canonical_record":{"source":{"id":"2605.19377","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T05:22:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"791a64563c30c11b64e7f92290ffdd42e1b9676b2f69028b4f8ab1c60858fdec","abstract_canon_sha256":"b6f65e463bca20fe0863be78ebdfc87bf6ea164b79f3dc0b0e861e5c86b2bbd6"},"schema_version":"1.0"},"canonical_sha256":"81ff88b8e34694c246f8e3f0081daeef5fe2efa4d845dcff0e40a84d584490f7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:42.627597Z","signature_b64":"DggC+3UvEiqPx+6IsfbYW8Z8bZj1FiMTEy5JIa9OWH8SYFWLIOZABFXBSLINBq1Shu0IjOHhMoQupl1bzFObCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81ff88b8e34694c246f8e3f0081daeef5fe2efa4d845dcff0e40a84d584490f7","last_reissued_at":"2026-05-20T01:05:42.626912Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:42.626912Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.19377","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T01:05:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CGBWeDBpfi7b+m3wZpk9sybPZR1QjLaUWZIMtzizvXnXDudoCOyA61iNyFB5Haz2QZenO5iDnmo8hMNclPhBCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T05:31:12.296390Z"},"content_sha256":"6594a1520351cb6a90b38c424b29964aa76e5d683e63b0a98bde57b442009204","schema_version":"1.0","event_id":"sha256:6594a1520351cb6a90b38c424b29964aa76e5d683e63b0a98bde57b442009204"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QH7YROHDI2KMERXY4PYAQHNO55","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Evaluation Game: Beyond Static LLM Benchmarking","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anne-Marie Kermarrec, Jade Garcia-Bourr\\'ee, Paul Wang, Vincent Corruble","submitted_at":"2026-05-19T05:22:21Z","abstract_excerpt":"As jailbreaks, adversarially crafted inputs that bypass safety constraints, continue to be discovered in Large Language Models, practitioners increasingly rely on fine-tuning as a defensive strategy. Yet the theoretical foundations underlying this robustness fine-tuning remain underexplored. We introduce a game-theoretic framework in which the interaction between an evaluator (auditing the model for jailbreaks) and a trainer is formalized as a two-player game.\n  A key feature of our approach is the use of group actions, a mathematical structure that captures symmetries and transformations, to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19377","kind":"arxiv","version":1},"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/2605.19377/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":"2026-05-20T01:05:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"U3c0YLSMr2+g0W7LHZ6HvvNn6JmppFoKihXgwRWp+t+aMJLQxz8XhIxMMS8L5i9PpzbQP2g2yWyFsn8l1Un3Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T05:31:12.296776Z"},"content_sha256":"f7c934182bca362d1769381f78fc3b6e5ebdf6a98dac3f1cf9fb423f4dc89293","schema_version":"1.0","event_id":"sha256:f7c934182bca362d1769381f78fc3b6e5ebdf6a98dac3f1cf9fb423f4dc89293"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:QH7YROHDI2KMERXY4PYAQHNO55","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.48550/arXiv.2512.13961.URLhttps://arxiv.org/abs/2512.13961) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Team Olmo. Olmo 3.arXiv preprint arXiv:2512.13961, 2025. doi: 10.48550/arXiv.2512. 13961. URLhttps://arxiv.org/abs/2512.13961","arxiv_id":"2605.19377","detector":"doi_compliance","evidence":{"ref_index":56,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Team Olmo. Olmo 3.arXiv preprint arXiv:2512.13961, 2025. doi: 10.48550/arXiv.2512. 13961. URLhttps://arxiv.org/abs/2512.13961","reconstructed_doi":"10.48550/arXiv.2512.13961.URLhttps://arxiv.org/abs/2512.13961"},"severity":"advisory","ref_index":56,"audited_at":"2026-05-20T08:00:08.573339Z","event_type":"pith.integrity.v1","detected_doi":"10.48550/arXiv.2512.13961.URLhttps://arxiv.org/abs/2512.13961","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"93e30e54771863a4cfd1881d3b8953ceea655826295761665fa4b8c5d20eb4a8","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":4909,"payload_sha256":"09bc9a81d32f6de87316e5064468956c0e915dbf1833b741764ce1520d61ebca","signature_b64":"D3KO8xtUbe7EcSe4hYcq2jCxR7uJh5ISz+QmUKyPQPxHalDOPDO/0ASjCSVhY3B5ML2Fvyw85NK2VXKUsTqoCA==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T08:02:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TsyrCMYHASrdrVLJbi+VupOaHMA21TGqcoWsXh333i3tUKOJ25O+AnHt4nu14CYy++/T07OHo3cwrKSdFqnFDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T05:31:12.297834Z"},"content_sha256":"9aa5ddef07190eee37c0abfd2de94a6d9352469167c7845d3c5911b7c3468653","schema_version":"1.0","event_id":"sha256:9aa5ddef07190eee37c0abfd2de94a6d9352469167c7845d3c5911b7c3468653"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QH7YROHDI2KMERXY4PYAQHNO55/bundle.json","state_url":"https://pith.science/pith/QH7YROHDI2KMERXY4PYAQHNO55/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QH7YROHDI2KMERXY4PYAQHNO55/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":"2026-05-21T05:31:12Z","links":{"resolver":"https://pith.science/pith/QH7YROHDI2KMERXY4PYAQHNO55","bundle":"https://pith.science/pith/QH7YROHDI2KMERXY4PYAQHNO55/bundle.json","state":"https://pith.science/pith/QH7YROHDI2KMERXY4PYAQHNO55/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QH7YROHDI2KMERXY4PYAQHNO55/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QH7YROHDI2KMERXY4PYAQHNO55","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b6f65e463bca20fe0863be78ebdfc87bf6ea164b79f3dc0b0e861e5c86b2bbd6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T05:22:21Z","title_canon_sha256":"791a64563c30c11b64e7f92290ffdd42e1b9676b2f69028b4f8ab1c60858fdec"},"schema_version":"1.0","source":{"id":"2605.19377","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19377","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19377v1","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19377","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_12","alias_value":"QH7YROHDI2KM","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_16","alias_value":"QH7YROHDI2KMERXY","created_at":"2026-05-20T01:05:42Z"},{"alias_kind":"pith_short_8","alias_value":"QH7YROHD","created_at":"2026-05-20T01:05:42Z"}],"graph_snapshots":[{"event_id":"sha256:f7c934182bca362d1769381f78fc3b6e5ebdf6a98dac3f1cf9fb423f4dc89293","target":"graph","created_at":"2026-05-20T01:05:42Z","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/2605.19377/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As jailbreaks, adversarially crafted inputs that bypass safety constraints, continue to be discovered in Large Language Models, practitioners increasingly rely on fine-tuning as a defensive strategy. Yet the theoretical foundations underlying this robustness fine-tuning remain underexplored. We introduce a game-theoretic framework in which the interaction between an evaluator (auditing the model for jailbreaks) and a trainer is formalized as a two-player game.\n  A key feature of our approach is the use of group actions, a mathematical structure that captures symmetries and transformations, to ","authors_text":"Anne-Marie Kermarrec, Jade Garcia-Bourr\\'ee, Paul Wang, Vincent Corruble","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T05:22:21Z","title":"The Evaluation Game: Beyond Static LLM Benchmarking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19377","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":"sha256:6594a1520351cb6a90b38c424b29964aa76e5d683e63b0a98bde57b442009204","target":"record","created_at":"2026-05-20T01:05:42Z","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":"b6f65e463bca20fe0863be78ebdfc87bf6ea164b79f3dc0b0e861e5c86b2bbd6","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-19T05:22:21Z","title_canon_sha256":"791a64563c30c11b64e7f92290ffdd42e1b9676b2f69028b4f8ab1c60858fdec"},"schema_version":"1.0","source":{"id":"2605.19377","kind":"arxiv","version":1}},"canonical_sha256":"81ff88b8e34694c246f8e3f0081daeef5fe2efa4d845dcff0e40a84d584490f7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"81ff88b8e34694c246f8e3f0081daeef5fe2efa4d845dcff0e40a84d584490f7","first_computed_at":"2026-05-20T01:05:42.626912Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:42.626912Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DggC+3UvEiqPx+6IsfbYW8Z8bZj1FiMTEy5JIa9OWH8SYFWLIOZABFXBSLINBq1Shu0IjOHhMoQupl1bzFObCQ==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:42.627597Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.19377","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6594a1520351cb6a90b38c424b29964aa76e5d683e63b0a98bde57b442009204","sha256:f7c934182bca362d1769381f78fc3b6e5ebdf6a98dac3f1cf9fb423f4dc89293","sha256:9aa5ddef07190eee37c0abfd2de94a6d9352469167c7845d3c5911b7c3468653"],"state_sha256":"17f9be9ea0519a4da53fa274a22c9f3571692c8ea6f8ac47ccd8d17ecc483f96"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"72Aon6qyCTpKXVoVA7ko2gStidVgOywxOH3XVJfjS+vCJ7vtcyPivIxPGMUeLOL7WMYGZNjwtjTkuuaX8GvzBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T05:31:12.299951Z","bundle_sha256":"9bbd7ab56954a792161499cd7a691d5599132c43d67a3b0e70d83f931eb3fcf3"}}