{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7TYY2XVBQ7LVZZEIFAYLCCSCH3","short_pith_number":"pith:7TYY2XVB","schema_version":"1.0","canonical_sha256":"fcf18d5ea187d75ce4882830b10a423ecc1619e2d1039186d67b605543cf4811","source":{"kind":"arxiv","id":"2505.18434","version":1},"attestation_state":"computed","paper":{"title":"TNG-CLIP:Training-Time Negation Data Generation for Negation Awareness of CLIP","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jesse Thomason, Mohammad Rostami, Yuliang Cai","submitted_at":"2025-05-24T00:02:48Z","abstract_excerpt":"Vision-language models (VLMs), such as CLIP, have demonstrated strong performance across a range of downstream tasks. However, CLIP is still limited in negation understanding: the ability to recognize the absence or exclusion of a concept. Existing methods address the problem by using a large language model (LLM) to generate large-scale data of image captions containing negation for further fine-tuning CLIP. However, these methods are both time- and compute-intensive, and their evaluations are typically restricted to image-text matching tasks. To expand the horizon, we (1) introduce a training"},"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":"2505.18434","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-24T00:02:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4b3de5715b8661295b5793f17428b34a25973c5465ed8e58035068e73f94ffb5","abstract_canon_sha256":"56b20aad305646df9c62ee3451a6e13de7878ef67abf55473db5a77402a71644"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:46.372934Z","signature_b64":"WhXvEsclxD/c+H1WcUSsPFbUzd58D2qW7N4nYp5GIxMMdmDvs12AxbDkGsQi88bUjAZkm24HhVcwf+GBX8hqAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcf18d5ea187d75ce4882830b10a423ecc1619e2d1039186d67b605543cf4811","last_reissued_at":"2026-07-05T11:08:46.372391Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:46.372391Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TNG-CLIP:Training-Time Negation Data Generation for Negation Awareness of CLIP","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jesse Thomason, Mohammad Rostami, Yuliang Cai","submitted_at":"2025-05-24T00:02:48Z","abstract_excerpt":"Vision-language models (VLMs), such as CLIP, have demonstrated strong performance across a range of downstream tasks. However, CLIP is still limited in negation understanding: the ability to recognize the absence or exclusion of a concept. Existing methods address the problem by using a large language model (LLM) to generate large-scale data of image captions containing negation for further fine-tuning CLIP. However, these methods are both time- and compute-intensive, and their evaluations are typically restricted to image-text matching tasks. To expand the horizon, we (1) introduce a training"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18434","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/2505.18434/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":"2505.18434","created_at":"2026-07-05T11:08:46.372455+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18434v1","created_at":"2026-07-05T11:08:46.372455+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18434","created_at":"2026-07-05T11:08:46.372455+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TYY2XVBQ7LV","created_at":"2026-07-05T11:08:46.372455+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TYY2XVBQ7LVZZEI","created_at":"2026-07-05T11:08:46.372455+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TYY2XVB","created_at":"2026-07-05T11:08:46.372455+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.00166","citing_title":"Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3","json":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3.json","graph_json":"https://pith.science/api/pith-number/7TYY2XVBQ7LVZZEIFAYLCCSCH3/graph.json","events_json":"https://pith.science/api/pith-number/7TYY2XVBQ7LVZZEIFAYLCCSCH3/events.json","paper":"https://pith.science/paper/7TYY2XVB"},"agent_actions":{"view_html":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3","download_json":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3.json","view_paper":"https://pith.science/paper/7TYY2XVB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18434&json=true","fetch_graph":"https://pith.science/api/pith-number/7TYY2XVBQ7LVZZEIFAYLCCSCH3/graph.json","fetch_events":"https://pith.science/api/pith-number/7TYY2XVBQ7LVZZEIFAYLCCSCH3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3/action/storage_attestation","attest_author":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3/action/author_attestation","sign_citation":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3/action/citation_signature","submit_replication":"https://pith.science/pith/7TYY2XVBQ7LVZZEIFAYLCCSCH3/action/replication_record"}},"created_at":"2026-07-05T11:08:46.372455+00:00","updated_at":"2026-07-05T11:08:46.372455+00:00"}