{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WSYMD6LDFSHAGF4ZORAJJCDBQZ","short_pith_number":"pith:WSYMD6LD","schema_version":"1.0","canonical_sha256":"b4b0c1f9632c8e0317997440948861865e94bf33fa27c82bde279fdcabb1c2a6","source":{"kind":"arxiv","id":"2507.20025","version":1},"attestation_state":"computed","paper":{"title":"Region-based Cluster Discrimination for Visual Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ismail Elezi, Jiankang Deng, Kaicheng Yang, Kun Wu, Roy Miles, Weimo Deng, Xiang An, Yin Xie, Yongle Zhao, Yumeng Wang, Zimin Ran, Ziyong Feng","submitted_at":"2025-07-26T17:47:09Z","abstract_excerpt":"Learning visual representations is foundational for a broad spectrum of downstream tasks. Although recent vision-language contrastive models, such as CLIP and SigLIP, have achieved impressive zero-shot performance via large-scale vision-language alignment, their reliance on global representations constrains their effectiveness for dense prediction tasks, such as grounding, OCR, and segmentation. To address this gap, we introduce Region-Aware Cluster Discrimination (RICE), a novel method that enhances region-level visual and OCR capabilities. We first construct a billion-scale candidate region "},"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":"2507.20025","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-26T17:47:09Z","cross_cats_sorted":[],"title_canon_sha256":"5eb9421ed10c7c467e882664c39f10f2c57878b56a64f159a292c878301fb95c","abstract_canon_sha256":"12a634875d3b8259da8e3412944a390f6c279f20d17d8a7dc522f6a637221ec8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:46.790731Z","signature_b64":"Ev/vdYtb84EwQ7gJb2Vg6bNziKM2EXiyTolnYVWiXuTiJHBKfKbV9k8wQ6JrfbCvgd6skywJ/ybyuEOoPv7NAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4b0c1f9632c8e0317997440948861865e94bf33fa27c82bde279fdcabb1c2a6","last_reissued_at":"2026-07-05T11:43:46.790173Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:46.790173Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Region-based Cluster Discrimination for Visual Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ismail Elezi, Jiankang Deng, Kaicheng Yang, Kun Wu, Roy Miles, Weimo Deng, Xiang An, Yin Xie, Yongle Zhao, Yumeng Wang, Zimin Ran, Ziyong Feng","submitted_at":"2025-07-26T17:47:09Z","abstract_excerpt":"Learning visual representations is foundational for a broad spectrum of downstream tasks. Although recent vision-language contrastive models, such as CLIP and SigLIP, have achieved impressive zero-shot performance via large-scale vision-language alignment, their reliance on global representations constrains their effectiveness for dense prediction tasks, such as grounding, OCR, and segmentation. To address this gap, we introduce Region-Aware Cluster Discrimination (RICE), a novel method that enhances region-level visual and OCR capabilities. We first construct a billion-scale candidate region "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.20025","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/2507.20025/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":"2507.20025","created_at":"2026-07-05T11:43:46.790239+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.20025v1","created_at":"2026-07-05T11:43:46.790239+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.20025","created_at":"2026-07-05T11:43:46.790239+00:00"},{"alias_kind":"pith_short_12","alias_value":"WSYMD6LDFSHA","created_at":"2026-07-05T11:43:46.790239+00:00"},{"alias_kind":"pith_short_16","alias_value":"WSYMD6LDFSHAGF4Z","created_at":"2026-07-05T11:43:46.790239+00:00"},{"alias_kind":"pith_short_8","alias_value":"WSYMD6LD","created_at":"2026-07-05T11:43:46.790239+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.16719","citing_title":"SAM 3: Segment Anything with Concepts","ref_index":142,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ","json":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ.json","graph_json":"https://pith.science/api/pith-number/WSYMD6LDFSHAGF4ZORAJJCDBQZ/graph.json","events_json":"https://pith.science/api/pith-number/WSYMD6LDFSHAGF4ZORAJJCDBQZ/events.json","paper":"https://pith.science/paper/WSYMD6LD"},"agent_actions":{"view_html":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ","download_json":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ.json","view_paper":"https://pith.science/paper/WSYMD6LD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.20025&json=true","fetch_graph":"https://pith.science/api/pith-number/WSYMD6LDFSHAGF4ZORAJJCDBQZ/graph.json","fetch_events":"https://pith.science/api/pith-number/WSYMD6LDFSHAGF4ZORAJJCDBQZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ/action/storage_attestation","attest_author":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ/action/author_attestation","sign_citation":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ/action/citation_signature","submit_replication":"https://pith.science/pith/WSYMD6LDFSHAGF4ZORAJJCDBQZ/action/replication_record"}},"created_at":"2026-07-05T11:43:46.790239+00:00","updated_at":"2026-07-05T11:43:46.790239+00:00"}