{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VBQEHYMUU3G6I4XYSNMTE7SLOT","short_pith_number":"pith:VBQEHYMU","schema_version":"1.0","canonical_sha256":"a86043e194a6cde472f89359327e4b74dd10e4b34bf1fbba9a4c3716491d9744","source":{"kind":"arxiv","id":"2404.01889","version":3},"attestation_state":"computed","paper":{"title":"RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Gregory Slabaugh, Martin Benning, Tatiana Gaintseva","submitted_at":"2024-04-02T12:28:40Z","abstract_excerpt":"In this paper we propose a novel modification of Contrastive Language-Image Pre-Training (CLIP) guidance for the task of unsupervised backlit image enhancement. Our work builds on the state-of-the-art CLIP-LIT approach, which learns a prompt pair by constraining the text-image similarity between a prompt (negative/positive sample) and a corresponding image (backlit image/well-lit image) in the CLIP embedding space. Learned prompts then guide an image enhancement network. Based on the CLIP-LIT framework, we propose two novel methods for CLIP guidance. First, we show that instead of tuning promp"},"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":"2404.01889","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-02T12:28:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e2a0396718dbc44f2a509d847a79c970f75a8468c9d5b515a9bf639e57164d5b","abstract_canon_sha256":"c82ba7107a1b10b0786e391e6e6cb50080fe3c1ac9eb503ff6c430071d2e3c9b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:46:23.624712Z","signature_b64":"J4jQAbzR8tgBA2kFuWJRL+K0GKkRjynFFPAu8yi4mgFZdHv++uDczMugr3gjstf8bqQDamT1BEvVU/BmdRiBBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a86043e194a6cde472f89359327e4b74dd10e4b34bf1fbba9a4c3716491d9744","last_reissued_at":"2026-07-05T08:46:23.624286Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:46:23.624286Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Gregory Slabaugh, Martin Benning, Tatiana Gaintseva","submitted_at":"2024-04-02T12:28:40Z","abstract_excerpt":"In this paper we propose a novel modification of Contrastive Language-Image Pre-Training (CLIP) guidance for the task of unsupervised backlit image enhancement. Our work builds on the state-of-the-art CLIP-LIT approach, which learns a prompt pair by constraining the text-image similarity between a prompt (negative/positive sample) and a corresponding image (backlit image/well-lit image) in the CLIP embedding space. Learned prompts then guide an image enhancement network. Based on the CLIP-LIT framework, we propose two novel methods for CLIP guidance. First, we show that instead of tuning promp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01889","kind":"arxiv","version":3},"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/2404.01889/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":"2404.01889","created_at":"2026-07-05T08:46:23.624348+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.01889v3","created_at":"2026-07-05T08:46:23.624348+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01889","created_at":"2026-07-05T08:46:23.624348+00:00"},{"alias_kind":"pith_short_12","alias_value":"VBQEHYMUU3G6","created_at":"2026-07-05T08:46:23.624348+00:00"},{"alias_kind":"pith_short_16","alias_value":"VBQEHYMUU3G6I4XY","created_at":"2026-07-05T08:46:23.624348+00:00"},{"alias_kind":"pith_short_8","alias_value":"VBQEHYMU","created_at":"2026-07-05T08:46:23.624348+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.13921","citing_title":"DIME-Net: A Dual-Illumination Adaptive Enhancement Network Based on Retinex and Mixture-of-Experts","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT","json":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT.json","graph_json":"https://pith.science/api/pith-number/VBQEHYMUU3G6I4XYSNMTE7SLOT/graph.json","events_json":"https://pith.science/api/pith-number/VBQEHYMUU3G6I4XYSNMTE7SLOT/events.json","paper":"https://pith.science/paper/VBQEHYMU"},"agent_actions":{"view_html":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT","download_json":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT.json","view_paper":"https://pith.science/paper/VBQEHYMU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.01889&json=true","fetch_graph":"https://pith.science/api/pith-number/VBQEHYMUU3G6I4XYSNMTE7SLOT/graph.json","fetch_events":"https://pith.science/api/pith-number/VBQEHYMUU3G6I4XYSNMTE7SLOT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT/action/storage_attestation","attest_author":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT/action/author_attestation","sign_citation":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT/action/citation_signature","submit_replication":"https://pith.science/pith/VBQEHYMUU3G6I4XYSNMTE7SLOT/action/replication_record"}},"created_at":"2026-07-05T08:46:23.624348+00:00","updated_at":"2026-07-05T08:46:23.624348+00:00"}