{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6HXEEPZ2N4L4TWNRLF2OIBOLHG","short_pith_number":"pith:6HXEEPZ2","schema_version":"1.0","canonical_sha256":"f1ee423f3a6f17c9d9b15974e405cb3982733085a2a5e77334cf77a1d934f777","source":{"kind":"arxiv","id":"2204.03647","version":1},"attestation_state":"computed","paper":{"title":"Adapting CLIP For Phrase Localization Without Further Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Greg Shakhnarovich, Jiahao Li, Raymond A. Yeh","submitted_at":"2022-04-07T17:59:38Z","abstract_excerpt":"Supervised or weakly supervised methods for phrase localization (textual grounding) either rely on human annotations or some other supervised models, e.g., object detectors. Obtaining these annotations is labor-intensive and may be difficult to scale in practice. We propose to leverage recent advances in contrastive language-vision models, CLIP, pre-trained on image and caption pairs collected from the internet. In its original form, CLIP only outputs an image-level embedding without any spatial resolution. We adapt CLIP to generate high-resolution spatial feature maps. Importantly, we can ext"},"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":"2204.03647","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-04-07T17:59:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"95dd9876129bdf4d1d92379d2b417dc9a27787d6e2730635588575e5369535e2","abstract_canon_sha256":"d77e0bb954cd3e72ba59425a8c69d3f1d38b2c91eb7061f50dca6631edb23449"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:32.619015Z","signature_b64":"DG1Dgy50TNLjeGqJP0YnH52VkYj4bBAeb/WGa9j8tLflJFf458u2iOZMkuStBFTbk9iaseY+VV7SqrFKp7JRAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1ee423f3a6f17c9d9b15974e405cb3982733085a2a5e77334cf77a1d934f777","last_reissued_at":"2026-07-05T04:12:32.618641Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:32.618641Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adapting CLIP For Phrase Localization Without Further Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Greg Shakhnarovich, Jiahao Li, Raymond A. Yeh","submitted_at":"2022-04-07T17:59:38Z","abstract_excerpt":"Supervised or weakly supervised methods for phrase localization (textual grounding) either rely on human annotations or some other supervised models, e.g., object detectors. Obtaining these annotations is labor-intensive and may be difficult to scale in practice. We propose to leverage recent advances in contrastive language-vision models, CLIP, pre-trained on image and caption pairs collected from the internet. In its original form, CLIP only outputs an image-level embedding without any spatial resolution. We adapt CLIP to generate high-resolution spatial feature maps. Importantly, we can ext"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.03647","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/2204.03647/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":"2204.03647","created_at":"2026-07-05T04:12:32.618701+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.03647v1","created_at":"2026-07-05T04:12:32.618701+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.03647","created_at":"2026-07-05T04:12:32.618701+00:00"},{"alias_kind":"pith_short_12","alias_value":"6HXEEPZ2N4L4","created_at":"2026-07-05T04:12:32.618701+00:00"},{"alias_kind":"pith_short_16","alias_value":"6HXEEPZ2N4L4TWNR","created_at":"2026-07-05T04:12:32.618701+00:00"},{"alias_kind":"pith_short_8","alias_value":"6HXEEPZ2","created_at":"2026-07-05T04:12:32.618701+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.09887","citing_title":"FLORA: Formal Language Model Enables Robust Training-free Zero-shot Object Referring Analysis","ref_index":75,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG","json":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG.json","graph_json":"https://pith.science/api/pith-number/6HXEEPZ2N4L4TWNRLF2OIBOLHG/graph.json","events_json":"https://pith.science/api/pith-number/6HXEEPZ2N4L4TWNRLF2OIBOLHG/events.json","paper":"https://pith.science/paper/6HXEEPZ2"},"agent_actions":{"view_html":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG","download_json":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG.json","view_paper":"https://pith.science/paper/6HXEEPZ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.03647&json=true","fetch_graph":"https://pith.science/api/pith-number/6HXEEPZ2N4L4TWNRLF2OIBOLHG/graph.json","fetch_events":"https://pith.science/api/pith-number/6HXEEPZ2N4L4TWNRLF2OIBOLHG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG/action/storage_attestation","attest_author":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG/action/author_attestation","sign_citation":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG/action/citation_signature","submit_replication":"https://pith.science/pith/6HXEEPZ2N4L4TWNRLF2OIBOLHG/action/replication_record"}},"created_at":"2026-07-05T04:12:32.618701+00:00","updated_at":"2026-07-05T04:12:32.618701+00:00"}