{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OGFSX3DWC34YJBWR4SGT2EXYFK","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":"001aaa0e65f2a0de61bfb0b8faf36b22cdd7b271f83429ab2dbfcf41b4ed0462","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T18:09:48Z","title_canon_sha256":"6568c0df4a83ea7ee1d81ec5d20c22b94b936b3bd1c4575ddc38be4f7ef62ab1"},"schema_version":"1.0","source":{"id":"2402.12336","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12336","created_at":"2026-07-05T08:27:46Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12336v2","created_at":"2026-07-05T08:27:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12336","created_at":"2026-07-05T08:27:46Z"},{"alias_kind":"pith_short_12","alias_value":"OGFSX3DWC34Y","created_at":"2026-07-05T08:27:46Z"},{"alias_kind":"pith_short_16","alias_value":"OGFSX3DWC34YJBWR","created_at":"2026-07-05T08:27:46Z"},{"alias_kind":"pith_short_8","alias_value":"OGFSX3DW","created_at":"2026-07-05T08:27:46Z"}],"graph_snapshots":[{"event_id":"sha256:2c2a1bc55db6fd0ccc8c95c8b393b3f0e899050c744c45f2bd1f0e3218b67a6b","target":"graph","created_at":"2026-07-05T08:27:46Z","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/2402.12336/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multi-modal foundation models like OpenFlamingo, LLaVA, and GPT-4 are increasingly used for various real-world tasks. Prior work has shown that these models are highly vulnerable to adversarial attacks on the vision modality. These attacks can be leveraged to spread fake information or defraud users, and thus pose a significant risk, which makes the robustness of large multi-modal foundation models a pressing problem. The CLIP model, or one of its variants, is used as a frozen vision encoder in many large vision-language models (LVLMs), e.g. LLaVA and OpenFlamingo. We propose an unsupervised a","authors_text":"Christian Schlarmann, Francesco Croce, Matthias Hein, Naman Deep Singh","cross_cats":["cs.AI","cs.CV","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T18:09:48Z","title":"Robust CLIP: Unsupervised Adversarial Fine-Tuning of Vision Embeddings for Robust Large Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12336","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":"sha256:181d7192eebbba81aae636bf7ddd70fa2f43f704653d2626fc1839039bc5c2d7","target":"record","created_at":"2026-07-05T08:27:46Z","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":"001aaa0e65f2a0de61bfb0b8faf36b22cdd7b271f83429ab2dbfcf41b4ed0462","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-19T18:09:48Z","title_canon_sha256":"6568c0df4a83ea7ee1d81ec5d20c22b94b936b3bd1c4575ddc38be4f7ef62ab1"},"schema_version":"1.0","source":{"id":"2402.12336","kind":"arxiv","version":2}},"canonical_sha256":"718b2bec7616f98486d1e48d3d12f82a85897243029a63382bc8663ce8c2a802","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"718b2bec7616f98486d1e48d3d12f82a85897243029a63382bc8663ce8c2a802","first_computed_at":"2026-07-05T08:27:46.372224Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:27:46.372224Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A5I/JScUJvKKPfKD9dPS3n00iRHTxwTe962rKYvU8POoB2/YDm21Sy0t00gbKpezwT7/gK1DmP9H53hrhy3XBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:27:46.372731Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.12336","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:181d7192eebbba81aae636bf7ddd70fa2f43f704653d2626fc1839039bc5c2d7","sha256:2c2a1bc55db6fd0ccc8c95c8b393b3f0e899050c744c45f2bd1f0e3218b67a6b"],"state_sha256":"80d4d7584f45f8c63759a2f8eec8b3bc9150e3edfdd2a55c4693c2e22f028119"}