{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YQHDL4I4SD3LMRQYXUEDCQPY7T","short_pith_number":"pith:YQHDL4I4","schema_version":"1.0","canonical_sha256":"c40e35f11c90f6b64618bd083141f8fcd0f9cc4cbfb42ac4c46bd73b278bf028","source":{"kind":"arxiv","id":"2312.00878","version":3},"attestation_state":"computed","paper":{"title":"Grounding Everything: Emerging Localization Properties in Vision-Language Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Felix Petersen, Hilde Kuehne, Vittorio Ferrari, Walid Bousselham","submitted_at":"2023-12-01T19:06:12Z","abstract_excerpt":"Vision-language foundation models have shown remarkable performance in various zero-shot settings such as image retrieval, classification, or captioning. But so far, those models seem to fall behind when it comes to zero-shot localization of referential expressions and objects in images. As a result, they need to be fine-tuned for this task. In this paper, we show that pretrained vision-language (VL) models allow for zero-shot open-vocabulary object localization without any fine-tuning. To leverage those capabilities, we propose a Grounding Everything Module (GEM) that generalizes the idea of "},"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":"2312.00878","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-12-01T19:06:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"880e01a7ea26b8144d232ac90f68f82ae9a84542d6eb980ab3a9d18717ddc022","abstract_canon_sha256":"6b3f318f83f1a956c191f263ac95c4cd449ccc19b20b0022135a79ed82a012fe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:24:02.555428Z","signature_b64":"1eCbTdKOy9hdFFbbZBbFqJTfbOCzdaiZJwuzmxtwPJQzKy6JuJxxI6GznTepXmr9tNL7HqPmN1tUac+crAOCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c40e35f11c90f6b64618bd083141f8fcd0f9cc4cbfb42ac4c46bd73b278bf028","last_reissued_at":"2026-07-05T07:24:02.554956Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:24:02.554956Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding Everything: Emerging Localization Properties in Vision-Language Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Felix Petersen, Hilde Kuehne, Vittorio Ferrari, Walid Bousselham","submitted_at":"2023-12-01T19:06:12Z","abstract_excerpt":"Vision-language foundation models have shown remarkable performance in various zero-shot settings such as image retrieval, classification, or captioning. But so far, those models seem to fall behind when it comes to zero-shot localization of referential expressions and objects in images. As a result, they need to be fine-tuned for this task. In this paper, we show that pretrained vision-language (VL) models allow for zero-shot open-vocabulary object localization without any fine-tuning. To leverage those capabilities, we propose a Grounding Everything Module (GEM) that generalizes the idea of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.00878","kind":"arxiv","version":3},"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/2312.00878/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":"2312.00878","created_at":"2026-07-05T07:24:02.555009+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.00878v3","created_at":"2026-07-05T07:24:02.555009+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.00878","created_at":"2026-07-05T07:24:02.555009+00:00"},{"alias_kind":"pith_short_12","alias_value":"YQHDL4I4SD3L","created_at":"2026-07-05T07:24:02.555009+00:00"},{"alias_kind":"pith_short_16","alias_value":"YQHDL4I4SD3LMRQY","created_at":"2026-07-05T07:24:02.555009+00:00"},{"alias_kind":"pith_short_8","alias_value":"YQHDL4I4","created_at":"2026-07-05T07:24:02.555009+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.23470","citing_title":"Interactive Interface For Semantic Segmentation Dataset Synthesis","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T","json":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T.json","graph_json":"https://pith.science/api/pith-number/YQHDL4I4SD3LMRQYXUEDCQPY7T/graph.json","events_json":"https://pith.science/api/pith-number/YQHDL4I4SD3LMRQYXUEDCQPY7T/events.json","paper":"https://pith.science/paper/YQHDL4I4"},"agent_actions":{"view_html":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T","download_json":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T.json","view_paper":"https://pith.science/paper/YQHDL4I4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.00878&json=true","fetch_graph":"https://pith.science/api/pith-number/YQHDL4I4SD3LMRQYXUEDCQPY7T/graph.json","fetch_events":"https://pith.science/api/pith-number/YQHDL4I4SD3LMRQYXUEDCQPY7T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T/action/storage_attestation","attest_author":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T/action/author_attestation","sign_citation":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T/action/citation_signature","submit_replication":"https://pith.science/pith/YQHDL4I4SD3LMRQYXUEDCQPY7T/action/replication_record"}},"created_at":"2026-07-05T07:24:02.555009+00:00","updated_at":"2026-07-05T07:24:02.555009+00:00"}