{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HSKLPDZ5BVS6YTZG7WZSR53I3U","short_pith_number":"pith:HSKLPDZ5","schema_version":"1.0","canonical_sha256":"3c94b78f3d0d65ec4f26fdb328f768dd3feba38d48897b51f650c553dea8aca1","source":{"kind":"arxiv","id":"2509.03494","version":2},"attestation_state":"computed","paper":{"title":"Parameter-Efficient Adaptation of mPLUG-Owl2 via Pixel-Level Visual Prompts for NR-IQA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mohammed EL Hassouni, Yahya Benmahane","submitted_at":"2025-09-03T17:23:24Z","abstract_excerpt":"In this paper, we propose a novel parameter-efficient adaptation method for No- Reference Image Quality Assessment (NR-IQA) using visual prompts optimized in pixel-space. Unlike full fine-tuning of Multimodal Large Language Models (MLLMs), our approach trains only 600K parameters at most (< 0.01% of the base model), while keeping the underlying model fully frozen. During inference, these visual prompts are combined with images via addition and processed by mPLUG-Owl2 with the textual query \"Rate the technical quality of the image.\" Evaluations across distortion types (synthetic, realistic, AI-"},"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":"2509.03494","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-03T17:23:24Z","cross_cats_sorted":[],"title_canon_sha256":"98ee8649d5379238cb8e3cdccecd92998b22624111e6cf83a2d4a8ac40e49369","abstract_canon_sha256":"861c3cbb93b4ea563d42ade4db6fa2aa0f562c3df0eef17a7a58b2139c57ec83"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:20.093580Z","signature_b64":"3QC1NQyPxFOPZwxTPURrw9F9bDVI/w0GwIT9wcPgiBE9DAROUAfQ+ONLrp3x6oV2Eudvy3lN1ZEZgsHC2Hb7Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c94b78f3d0d65ec4f26fdb328f768dd3feba38d48897b51f650c553dea8aca1","last_reissued_at":"2026-07-05T12:06:20.093014Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:20.093014Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Parameter-Efficient Adaptation of mPLUG-Owl2 via Pixel-Level Visual Prompts for NR-IQA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Mohammed EL Hassouni, Yahya Benmahane","submitted_at":"2025-09-03T17:23:24Z","abstract_excerpt":"In this paper, we propose a novel parameter-efficient adaptation method for No- Reference Image Quality Assessment (NR-IQA) using visual prompts optimized in pixel-space. Unlike full fine-tuning of Multimodal Large Language Models (MLLMs), our approach trains only 600K parameters at most (< 0.01% of the base model), while keeping the underlying model fully frozen. During inference, these visual prompts are combined with images via addition and processed by mPLUG-Owl2 with the textual query \"Rate the technical quality of the image.\" Evaluations across distortion types (synthetic, realistic, AI-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.03494","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/2509.03494/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":"2509.03494","created_at":"2026-07-05T12:06:20.093069+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.03494v2","created_at":"2026-07-05T12:06:20.093069+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.03494","created_at":"2026-07-05T12:06:20.093069+00:00"},{"alias_kind":"pith_short_12","alias_value":"HSKLPDZ5BVS6","created_at":"2026-07-05T12:06:20.093069+00:00"},{"alias_kind":"pith_short_16","alias_value":"HSKLPDZ5BVS6YTZG","created_at":"2026-07-05T12:06:20.093069+00:00"},{"alias_kind":"pith_short_8","alias_value":"HSKLPDZ5","created_at":"2026-07-05T12:06:20.093069+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/HSKLPDZ5BVS6YTZG7WZSR53I3U","json":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U.json","graph_json":"https://pith.science/api/pith-number/HSKLPDZ5BVS6YTZG7WZSR53I3U/graph.json","events_json":"https://pith.science/api/pith-number/HSKLPDZ5BVS6YTZG7WZSR53I3U/events.json","paper":"https://pith.science/paper/HSKLPDZ5"},"agent_actions":{"view_html":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U","download_json":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U.json","view_paper":"https://pith.science/paper/HSKLPDZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.03494&json=true","fetch_graph":"https://pith.science/api/pith-number/HSKLPDZ5BVS6YTZG7WZSR53I3U/graph.json","fetch_events":"https://pith.science/api/pith-number/HSKLPDZ5BVS6YTZG7WZSR53I3U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U/action/storage_attestation","attest_author":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U/action/author_attestation","sign_citation":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U/action/citation_signature","submit_replication":"https://pith.science/pith/HSKLPDZ5BVS6YTZG7WZSR53I3U/action/replication_record"}},"created_at":"2026-07-05T12:06:20.093069+00:00","updated_at":"2026-07-05T12:06:20.093069+00:00"}