{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JFKT5Y3MZWHEZX4OFCWKUDDRNU","short_pith_number":"pith:JFKT5Y3M","schema_version":"1.0","canonical_sha256":"49553ee36ccd8e4cdf8e28acaa0c716d0755088a42c0c72de526c0670b651ef3","source":{"kind":"arxiv","id":"2502.01576","version":2},"attestation_state":"computed","paper":{"title":"Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fahad Khan, Fahad Shamshad, Hashmat Shadab Malik, Karthik Nandakumar, Muzammal Naseer, Salman Khan","submitted_at":"2025-02-03T17:59:45Z","abstract_excerpt":"Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate responses, or bypass safety mechanisms. Existing methods seek to mitigate these risks by applying constrained adversarial fine-tuning to CLIP vision encoders on ImageNet-scale data, ensuring their generalization ability is preserved. However, this limited adversarial training restricts robustness and broader generalization. In this work, we explore an alternative approach of leveraging existing vision classification models"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2502.01576","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-03T17:59:45Z","cross_cats_sorted":[],"title_canon_sha256":"8a8ad3f38ed1f19bf79686e2907a9b69ba45b778aaa05fe0e13ae307ca756865","abstract_canon_sha256":"0fb6b218068a6eed449b7e1edc564354618dfb9844cc9752ffd788826c5a188d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:27.517294Z","signature_b64":"W8n/ti3zY3ma+dRuJcXn3fhu7h1UFkTgbTsosoqtb/PqVvoIuEGYd2T2ZyPrNqD71ufAxcTFLc48pVv8Bmc9AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"49553ee36ccd8e4cdf8e28acaa0c716d0755088a42c0c72de526c0670b651ef3","last_reissued_at":"2026-06-04T01:08:27.516696Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:27.516696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust-LLaVA: On the Effectiveness of Large-Scale Robust Image Encoders for Multi-modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fahad Khan, Fahad Shamshad, Hashmat Shadab Malik, Karthik Nandakumar, Muzammal Naseer, Salman Khan","submitted_at":"2025-02-03T17:59:45Z","abstract_excerpt":"Multi-modal Large Language Models (MLLMs) excel in vision-language tasks but remain vulnerable to visual adversarial perturbations that can induce hallucinations, manipulate responses, or bypass safety mechanisms. Existing methods seek to mitigate these risks by applying constrained adversarial fine-tuning to CLIP vision encoders on ImageNet-scale data, ensuring their generalization ability is preserved. However, this limited adversarial training restricts robustness and broader generalization. In this work, we explore an alternative approach of leveraging existing vision classification models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01576","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/2502.01576/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2502.01576","created_at":"2026-06-04T01:08:27.516779+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.01576v2","created_at":"2026-06-04T01:08:27.516779+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01576","created_at":"2026-06-04T01:08:27.516779+00:00"},{"alias_kind":"pith_short_12","alias_value":"JFKT5Y3MZWHE","created_at":"2026-06-04T01:08:27.516779+00:00"},{"alias_kind":"pith_short_16","alias_value":"JFKT5Y3MZWHEZX4O","created_at":"2026-06-04T01:08:27.516779+00:00"},{"alias_kind":"pith_short_8","alias_value":"JFKT5Y3M","created_at":"2026-06-04T01:08:27.516779+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU","json":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU.json","graph_json":"https://pith.science/api/pith-number/JFKT5Y3MZWHEZX4OFCWKUDDRNU/graph.json","events_json":"https://pith.science/api/pith-number/JFKT5Y3MZWHEZX4OFCWKUDDRNU/events.json","paper":"https://pith.science/paper/JFKT5Y3M"},"agent_actions":{"view_html":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU","download_json":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU.json","view_paper":"https://pith.science/paper/JFKT5Y3M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.01576&json=true","fetch_graph":"https://pith.science/api/pith-number/JFKT5Y3MZWHEZX4OFCWKUDDRNU/graph.json","fetch_events":"https://pith.science/api/pith-number/JFKT5Y3MZWHEZX4OFCWKUDDRNU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU/action/storage_attestation","attest_author":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU/action/author_attestation","sign_citation":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU/action/citation_signature","submit_replication":"https://pith.science/pith/JFKT5Y3MZWHEZX4OFCWKUDDRNU/action/replication_record"}},"created_at":"2026-06-04T01:08:27.516779+00:00","updated_at":"2026-06-04T01:08:27.516779+00:00"}