{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Z4ZJM6S5VHHJBAJMB5ZUP3KW5L","short_pith_number":"pith:Z4ZJM6S5","schema_version":"1.0","canonical_sha256":"cf32967a5da9ce90812c0f7347ed56eaddac9d86e0f0d3bbc3b72ed888b4e652","source":{"kind":"arxiv","id":"2507.03458","version":1},"attestation_state":"computed","paper":{"title":"Helping CLIP See Both the Forest and the Trees: A Decomposition and Description Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changqing Zhang, Guangyu Wang, Leyan Xue, Mingyue Cheng, Qinghua Hu, Zongbo Han","submitted_at":"2025-07-04T10:24:26Z","abstract_excerpt":"Vision-Language Models (VLMs) like CLIP achieve cross-modal semantic alignment through contrastive learning, exhibiting robust zero-shot generalization. Traditional prompt engineering, however, predominantly relies on coarse-grained category labels, neglecting fine-grained local semantics. Existing approaches assume that VLMs inherently recognize localized visual details and attempt to enhance classification by augmenting text prompts with attribute descriptors generated by large language models. However, our systematic experiments reveal critical limitations: CLIP's strong bias toward global "},"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":"2507.03458","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-04T10:24:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e6e09ad158b1c893cbb3b783e59d29da30082176f786a78d7c12c96ae6f22b0b","abstract_canon_sha256":"87825de7961b08ab09bc06d9e276226b002611e5e5c3666075a749f652d4e7bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:59.942758Z","signature_b64":"0djL6iyPNv748tRjG6A6qbveZ91Sw/3bkiZpsyxLzmUfe2tNxwUPa73IsfCZRIgxvBdTEvJ1iBUDDYJ+hMLcCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf32967a5da9ce90812c0f7347ed56eaddac9d86e0f0d3bbc3b72ed888b4e652","last_reissued_at":"2026-07-05T11:31:59.942310Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:59.942310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Helping CLIP See Both the Forest and the Trees: A Decomposition and Description Approach","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changqing Zhang, Guangyu Wang, Leyan Xue, Mingyue Cheng, Qinghua Hu, Zongbo Han","submitted_at":"2025-07-04T10:24:26Z","abstract_excerpt":"Vision-Language Models (VLMs) like CLIP achieve cross-modal semantic alignment through contrastive learning, exhibiting robust zero-shot generalization. Traditional prompt engineering, however, predominantly relies on coarse-grained category labels, neglecting fine-grained local semantics. Existing approaches assume that VLMs inherently recognize localized visual details and attempt to enhance classification by augmenting text prompts with attribute descriptors generated by large language models. However, our systematic experiments reveal critical limitations: CLIP's strong bias toward global "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03458","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/2507.03458/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":"2507.03458","created_at":"2026-07-05T11:31:59.942380+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.03458v1","created_at":"2026-07-05T11:31:59.942380+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03458","created_at":"2026-07-05T11:31:59.942380+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z4ZJM6S5VHHJ","created_at":"2026-07-05T11:31:59.942380+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z4ZJM6S5VHHJBAJM","created_at":"2026-07-05T11:31:59.942380+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z4ZJM6S5","created_at":"2026-07-05T11:31:59.942380+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/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L","json":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L.json","graph_json":"https://pith.science/api/pith-number/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/graph.json","events_json":"https://pith.science/api/pith-number/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/events.json","paper":"https://pith.science/paper/Z4ZJM6S5"},"agent_actions":{"view_html":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L","download_json":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L.json","view_paper":"https://pith.science/paper/Z4ZJM6S5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.03458&json=true","fetch_graph":"https://pith.science/api/pith-number/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/graph.json","fetch_events":"https://pith.science/api/pith-number/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/action/storage_attestation","attest_author":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/action/author_attestation","sign_citation":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/action/citation_signature","submit_replication":"https://pith.science/pith/Z4ZJM6S5VHHJBAJMB5ZUP3KW5L/action/replication_record"}},"created_at":"2026-07-05T11:31:59.942380+00:00","updated_at":"2026-07-05T11:31:59.942380+00:00"}