{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MZAA6UVSVAVHAAUSCPN7DGU5DL","short_pith_number":"pith:MZAA6UVS","schema_version":"1.0","canonical_sha256":"66400f52b2a82a70029213dbf19a9d1ad5901d37d4cd6c039f2b896bd83c4f0c","source":{"kind":"arxiv","id":"2401.05224","version":2},"attestation_state":"computed","paper":{"title":"Do Vision and Language Encoders Represent the World Similarly?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Karttikeya Mangalam, Mayug Maniparambil, Mohamed El Amine Seddik, Noel E. O'Connor, Raiymbek Akshulakov, Sanath Narayan, Yasser Abdelaziz Dahou Djilali","submitted_at":"2024-01-10T15:51:39Z","abstract_excerpt":"Aligned text-image encoders such as CLIP have become the de facto model for vision-language tasks. Furthermore, modality-specific encoders achieve impressive performances in their respective domains. This raises a central question: does an alignment exist between uni-modal vision and language encoders since they fundamentally represent the same physical world? Analyzing the latent spaces structure of vision and language models on image-caption benchmarks using the Centered Kernel Alignment (CKA), we find that the representation spaces of unaligned and aligned encoders are semantically similar."},"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":"2401.05224","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-10T15:51:39Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"e1014355693289cae6c77dbbf0758f3ea5816fb811af76181418c6bd7fc077e5","abstract_canon_sha256":"8358451cfc449d17c8b702cf8650c93b36dc0dad1ef5fe52bbd2573958e665ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:59:36.106792Z","signature_b64":"XmaPFnFF0BtJyyTifKyOrOY+l6S+VZ+tUhjkLxtVzcHANHsyGX3j1VafMu0FIR6kZ764W55k09HMbABu9RtKCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"66400f52b2a82a70029213dbf19a9d1ad5901d37d4cd6c039f2b896bd83c4f0c","last_reissued_at":"2026-07-05T07:59:36.106301Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:59:36.106301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Vision and Language Encoders Represent the World Similarly?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Karttikeya Mangalam, Mayug Maniparambil, Mohamed El Amine Seddik, Noel E. O'Connor, Raiymbek Akshulakov, Sanath Narayan, Yasser Abdelaziz Dahou Djilali","submitted_at":"2024-01-10T15:51:39Z","abstract_excerpt":"Aligned text-image encoders such as CLIP have become the de facto model for vision-language tasks. Furthermore, modality-specific encoders achieve impressive performances in their respective domains. This raises a central question: does an alignment exist between uni-modal vision and language encoders since they fundamentally represent the same physical world? Analyzing the latent spaces structure of vision and language models on image-caption benchmarks using the Centered Kernel Alignment (CKA), we find that the representation spaces of unaligned and aligned encoders are semantically similar."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.05224","kind":"arxiv","version":2},"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/2401.05224/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":"2401.05224","created_at":"2026-07-05T07:59:36.106369+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.05224v2","created_at":"2026-07-05T07:59:36.106369+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.05224","created_at":"2026-07-05T07:59:36.106369+00:00"},{"alias_kind":"pith_short_12","alias_value":"MZAA6UVSVAVH","created_at":"2026-07-05T07:59:36.106369+00:00"},{"alias_kind":"pith_short_16","alias_value":"MZAA6UVSVAVHAAUS","created_at":"2026-07-05T07:59:36.106369+00:00"},{"alias_kind":"pith_short_8","alias_value":"MZAA6UVS","created_at":"2026-07-05T07:59:36.106369+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/MZAA6UVSVAVHAAUSCPN7DGU5DL","json":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL.json","graph_json":"https://pith.science/api/pith-number/MZAA6UVSVAVHAAUSCPN7DGU5DL/graph.json","events_json":"https://pith.science/api/pith-number/MZAA6UVSVAVHAAUSCPN7DGU5DL/events.json","paper":"https://pith.science/paper/MZAA6UVS"},"agent_actions":{"view_html":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL","download_json":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL.json","view_paper":"https://pith.science/paper/MZAA6UVS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.05224&json=true","fetch_graph":"https://pith.science/api/pith-number/MZAA6UVSVAVHAAUSCPN7DGU5DL/graph.json","fetch_events":"https://pith.science/api/pith-number/MZAA6UVSVAVHAAUSCPN7DGU5DL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL/action/storage_attestation","attest_author":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL/action/author_attestation","sign_citation":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL/action/citation_signature","submit_replication":"https://pith.science/pith/MZAA6UVSVAVHAAUSCPN7DGU5DL/action/replication_record"}},"created_at":"2026-07-05T07:59:36.106369+00:00","updated_at":"2026-07-05T07:59:36.106369+00:00"}