{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:F6ZJZXMDGRMCFGFFL4PUPBV3H7","short_pith_number":"pith:F6ZJZXMD","schema_version":"1.0","canonical_sha256":"2fb29cdd8334582298a55f1f4786bb3fde9958a4b0161c94057926b6940a1f78","source":{"kind":"arxiv","id":"2606.04180","version":1},"attestation_state":"computed","paper":{"title":"KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Farzan Farnia, Mohammad Jalali, Youqi Wu","submitted_at":"2026-06-02T19:47:41Z","abstract_excerpt":"Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems. While these models are typically compared through downstream performance, such evaluations often do not explain how their representations differ structurally. In this work, we study this problem through the task of Contrastive Embedding Clustering: identifying sample subsets that are weakly clustered under one representation but strongly clustered under another. We propose \\emph{Kernel Optimization for Discrepancy Analysis (KODA)}, a kernel-based framework for contrast"},"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":"2606.04180","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-02T19:47:41Z","cross_cats_sorted":["cs.IT","math.IT"],"title_canon_sha256":"26fe1514575cb5ce6440e17394179db6fecea196f3adfa8604fe43546cc3c7b1","abstract_canon_sha256":"3c52378fd54b743a16d2a5a765bad258e1228fdf3d9e42f8922a66df1d711e7e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:08:19.224280Z","signature_b64":"GaCaI4uCkA5mjNwHnwz17wbTUdDWbuanHu+65nydG/kqJk5Gj9CwwAL31YR1DlNG1Y5T9/HXjfR2YHx7BzgMAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fb29cdd8334582298a55f1f4786bb3fde9958a4b0161c94057926b6940a1f78","last_reissued_at":"2026-06-04T01:08:19.223626Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:08:19.223626Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Farzan Farnia, Mohammad Jalali, Youqi Wu","submitted_at":"2026-06-02T19:47:41Z","abstract_excerpt":"Vision-language foundation models such as CLIP and SigLIP provide widely used representations for multimodal learning systems. While these models are typically compared through downstream performance, such evaluations often do not explain how their representations differ structurally. In this work, we study this problem through the task of Contrastive Embedding Clustering: identifying sample subsets that are weakly clustered under one representation but strongly clustered under another. We propose \\emph{Kernel Optimization for Discrepancy Analysis (KODA)}, a kernel-based framework for contrast"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04180","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/2606.04180/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":"2606.04180","created_at":"2026-06-04T01:08:19.223730+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04180v1","created_at":"2026-06-04T01:08:19.223730+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04180","created_at":"2026-06-04T01:08:19.223730+00:00"},{"alias_kind":"pith_short_12","alias_value":"F6ZJZXMDGRMC","created_at":"2026-06-04T01:08:19.223730+00:00"},{"alias_kind":"pith_short_16","alias_value":"F6ZJZXMDGRMCFGFF","created_at":"2026-06-04T01:08:19.223730+00:00"},{"alias_kind":"pith_short_8","alias_value":"F6ZJZXMD","created_at":"2026-06-04T01:08:19.223730+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/F6ZJZXMDGRMCFGFFL4PUPBV3H7","json":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7.json","graph_json":"https://pith.science/api/pith-number/F6ZJZXMDGRMCFGFFL4PUPBV3H7/graph.json","events_json":"https://pith.science/api/pith-number/F6ZJZXMDGRMCFGFFL4PUPBV3H7/events.json","paper":"https://pith.science/paper/F6ZJZXMD"},"agent_actions":{"view_html":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7","download_json":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7.json","view_paper":"https://pith.science/paper/F6ZJZXMD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04180&json=true","fetch_graph":"https://pith.science/api/pith-number/F6ZJZXMDGRMCFGFFL4PUPBV3H7/graph.json","fetch_events":"https://pith.science/api/pith-number/F6ZJZXMDGRMCFGFFL4PUPBV3H7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7/action/storage_attestation","attest_author":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7/action/author_attestation","sign_citation":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7/action/citation_signature","submit_replication":"https://pith.science/pith/F6ZJZXMDGRMCFGFFL4PUPBV3H7/action/replication_record"}},"created_at":"2026-06-04T01:08:19.223730+00:00","updated_at":"2026-06-04T01:08:19.223730+00:00"}