{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RPXPLSEUUV72XYJRT5VWD3GEDE","short_pith_number":"pith:RPXPLSEU","schema_version":"1.0","canonical_sha256":"8beef5c894a57fabe1319f6b61ecc41921eae87d9bb78ed84409bc2e177e8dda","source":{"kind":"arxiv","id":"2505.13233","version":1},"attestation_state":"computed","paper":{"title":"From Local Details to Global Context: Advancing Vision-Language Models with Attention-Based Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Binhui Xie, Jian Liang, Jingxuan Kang, Lincan Cai, Shuang Li, Wenxuan Ma, Zhida Qin","submitted_at":"2025-05-19T15:15:37Z","abstract_excerpt":"Pretrained vision-language models (VLMs), e.g., CLIP, demonstrate impressive zero-shot capabilities on downstream tasks. Prior research highlights the crucial role of visual augmentation techniques, like random cropping, in alignment with fine-grained class descriptions generated by large language models (LLMs), significantly enhancing zero-shot performance by incorporating multi-view information. However, the inherent randomness of these augmentations can inevitably introduce background artifacts and cause models to overly focus on local details, compromising global semantic understanding. To"},"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":"2505.13233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-19T15:15:37Z","cross_cats_sorted":[],"title_canon_sha256":"3f0c37fcaee52aedc2632142db097b5cd4b6596be6aa285b5dddc73027cb4c93","abstract_canon_sha256":"e4a269adbda2db8557fa3d7ada1aaccbb12cc1aad9ef8e457b607bbabf5ae0dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:27.136225Z","signature_b64":"jAGNnoFY8eheS9ilC3ik4GOmqvTqaoLHSZL6PAqM1rtwU9jMflwzzGaeRuvx8djZDRG/FTOkhjy4Fk7Dn63aDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8beef5c894a57fabe1319f6b61ecc41921eae87d9bb78ed84409bc2e177e8dda","last_reissued_at":"2026-07-05T11:05:27.135719Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:27.135719Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Local Details to Global Context: Advancing Vision-Language Models with Attention-Based Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Binhui Xie, Jian Liang, Jingxuan Kang, Lincan Cai, Shuang Li, Wenxuan Ma, Zhida Qin","submitted_at":"2025-05-19T15:15:37Z","abstract_excerpt":"Pretrained vision-language models (VLMs), e.g., CLIP, demonstrate impressive zero-shot capabilities on downstream tasks. Prior research highlights the crucial role of visual augmentation techniques, like random cropping, in alignment with fine-grained class descriptions generated by large language models (LLMs), significantly enhancing zero-shot performance by incorporating multi-view information. However, the inherent randomness of these augmentations can inevitably introduce background artifacts and cause models to overly focus on local details, compromising global semantic understanding. To"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.13233","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/2505.13233/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":"2505.13233","created_at":"2026-07-05T11:05:27.135780+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.13233v1","created_at":"2026-07-05T11:05:27.135780+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.13233","created_at":"2026-07-05T11:05:27.135780+00:00"},{"alias_kind":"pith_short_12","alias_value":"RPXPLSEUUV72","created_at":"2026-07-05T11:05:27.135780+00:00"},{"alias_kind":"pith_short_16","alias_value":"RPXPLSEUUV72XYJR","created_at":"2026-07-05T11:05:27.135780+00:00"},{"alias_kind":"pith_short_8","alias_value":"RPXPLSEU","created_at":"2026-07-05T11:05:27.135780+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/RPXPLSEUUV72XYJRT5VWD3GEDE","json":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE.json","graph_json":"https://pith.science/api/pith-number/RPXPLSEUUV72XYJRT5VWD3GEDE/graph.json","events_json":"https://pith.science/api/pith-number/RPXPLSEUUV72XYJRT5VWD3GEDE/events.json","paper":"https://pith.science/paper/RPXPLSEU"},"agent_actions":{"view_html":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE","download_json":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE.json","view_paper":"https://pith.science/paper/RPXPLSEU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.13233&json=true","fetch_graph":"https://pith.science/api/pith-number/RPXPLSEUUV72XYJRT5VWD3GEDE/graph.json","fetch_events":"https://pith.science/api/pith-number/RPXPLSEUUV72XYJRT5VWD3GEDE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE/action/storage_attestation","attest_author":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE/action/author_attestation","sign_citation":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE/action/citation_signature","submit_replication":"https://pith.science/pith/RPXPLSEUUV72XYJRT5VWD3GEDE/action/replication_record"}},"created_at":"2026-07-05T11:05:27.135780+00:00","updated_at":"2026-07-05T11:05:27.135780+00:00"}