{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:6KS2H34VN2SN5OIGUE3AL36VSO","short_pith_number":"pith:6KS2H34V","canonical_record":{"source":{"id":"2412.01814","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:56:06Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"3a507b98d9203407403c3cb4bd28fe0b7b87648985acef12bdc87bba90b1b695","abstract_canon_sha256":"922323f469886f120358a5c487a7e5a05ae6290053ba45122545c13f04c2fdcf"},"schema_version":"1.0"},"canonical_sha256":"f2a5a3ef956ea4deb906a13605efd593a56e97fd302dede3c5512dbd761269b9","source":{"kind":"arxiv","id":"2412.01814","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01814","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01814v2","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01814","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_12","alias_value":"6KS2H34VN2SN","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_16","alias_value":"6KS2H34VN2SN5OIG","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_8","alias_value":"6KS2H34V","created_at":"2026-07-05T10:39:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:6KS2H34VN2SN5OIGUE3AL36VSO","target":"record","payload":{"canonical_record":{"source":{"id":"2412.01814","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:56:06Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"3a507b98d9203407403c3cb4bd28fe0b7b87648985acef12bdc87bba90b1b695","abstract_canon_sha256":"922323f469886f120358a5c487a7e5a05ae6290053ba45122545c13f04c2fdcf"},"schema_version":"1.0"},"canonical_sha256":"f2a5a3ef956ea4deb906a13605efd593a56e97fd302dede3c5512dbd761269b9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:23.753025Z","signature_b64":"Hpc3Z7dCK2mPwm5tOpTVRugJsGYIken2wh+FkC7xUr2G6f1gi3bJGaMgrY4BVqTiampMTgPlP6BxBkMF2GiPBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2a5a3ef956ea4deb906a13605efd593a56e97fd302dede3c5512dbd761269b9","last_reissued_at":"2026-07-05T10:39:23.752460Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:23.752460Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.01814","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:39:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zCvkG49Fahu/H/0X0zVdOuUtrX1hefKNsQPR+c+20/3IwR16mLVEIVMx6ApsdYJ7zBzga8ye67mAZc0/dWBXCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T16:52:09.586361Z"},"content_sha256":"69f8d4e807445d1921e7c808239d7b025474f54ee714651985e926b4b4d43708","schema_version":"1.0","event_id":"sha256:69f8d4e807445d1921e7c808239d7b025474f54ee714651985e926b4b4d43708"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:6KS2H34VN2SN5OIGUE3AL36VSO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Mariana-Iuliana Georgescu, Rui Xiao, Sanghwan Kim, Stephan Alaniz, Zeynep Akata","submitted_at":"2024-12-02T18:56:06Z","abstract_excerpt":"Vision-Language Models (VLMs) trained with contrastive loss have achieved significant advancements in various vision and language tasks. However, the global nature of the contrastive loss makes VLMs focus predominantly on foreground objects, neglecting other crucial information in the image, which limits their effectiveness in downstream tasks. To address these challenges, we propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training that integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework. We create global and l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01814","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/2412.01814/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:39:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PlU+zYqD9LQ8Y80LfKQlqXjpE7qJhiA0zhBcIaY34KjF53fOCvx5H9BjPepiCnKt3v6uljh0iyfnJgenW/1ZCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T16:52:09.586668Z"},"content_sha256":"662d55f4ed83d3628751c79af84b2155e41175cc65b3fdce9f5f51847d5e3b82","schema_version":"1.0","event_id":"sha256:662d55f4ed83d3628751c79af84b2155e41175cc65b3fdce9f5f51847d5e3b82"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6KS2H34VN2SN5OIGUE3AL36VSO/bundle.json","state_url":"https://pith.science/pith/6KS2H34VN2SN5OIGUE3AL36VSO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6KS2H34VN2SN5OIGUE3AL36VSO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-31T16:52:09Z","links":{"resolver":"https://pith.science/pith/6KS2H34VN2SN5OIGUE3AL36VSO","bundle":"https://pith.science/pith/6KS2H34VN2SN5OIGUE3AL36VSO/bundle.json","state":"https://pith.science/pith/6KS2H34VN2SN5OIGUE3AL36VSO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6KS2H34VN2SN5OIGUE3AL36VSO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6KS2H34VN2SN5OIGUE3AL36VSO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"922323f469886f120358a5c487a7e5a05ae6290053ba45122545c13f04c2fdcf","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:56:06Z","title_canon_sha256":"3a507b98d9203407403c3cb4bd28fe0b7b87648985acef12bdc87bba90b1b695"},"schema_version":"1.0","source":{"id":"2412.01814","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01814","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01814v2","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01814","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_12","alias_value":"6KS2H34VN2SN","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_16","alias_value":"6KS2H34VN2SN5OIG","created_at":"2026-07-05T10:39:23Z"},{"alias_kind":"pith_short_8","alias_value":"6KS2H34V","created_at":"2026-07-05T10:39:23Z"}],"graph_snapshots":[{"event_id":"sha256:662d55f4ed83d3628751c79af84b2155e41175cc65b3fdce9f5f51847d5e3b82","target":"graph","created_at":"2026-07-05T10:39:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.01814/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vision-Language Models (VLMs) trained with contrastive loss have achieved significant advancements in various vision and language tasks. However, the global nature of the contrastive loss makes VLMs focus predominantly on foreground objects, neglecting other crucial information in the image, which limits their effectiveness in downstream tasks. To address these challenges, we propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training that integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework. We create global and l","authors_text":"Mariana-Iuliana Georgescu, Rui Xiao, Sanghwan Kim, Stephan Alaniz, Zeynep Akata","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:56:06Z","title":"COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01814","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:69f8d4e807445d1921e7c808239d7b025474f54ee714651985e926b4b4d43708","target":"record","created_at":"2026-07-05T10:39:23Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"922323f469886f120358a5c487a7e5a05ae6290053ba45122545c13f04c2fdcf","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:56:06Z","title_canon_sha256":"3a507b98d9203407403c3cb4bd28fe0b7b87648985acef12bdc87bba90b1b695"},"schema_version":"1.0","source":{"id":"2412.01814","kind":"arxiv","version":2}},"canonical_sha256":"f2a5a3ef956ea4deb906a13605efd593a56e97fd302dede3c5512dbd761269b9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f2a5a3ef956ea4deb906a13605efd593a56e97fd302dede3c5512dbd761269b9","first_computed_at":"2026-07-05T10:39:23.752460Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:39:23.752460Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Hpc3Z7dCK2mPwm5tOpTVRugJsGYIken2wh+FkC7xUr2G6f1gi3bJGaMgrY4BVqTiampMTgPlP6BxBkMF2GiPBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:39:23.753025Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.01814","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:69f8d4e807445d1921e7c808239d7b025474f54ee714651985e926b4b4d43708","sha256:662d55f4ed83d3628751c79af84b2155e41175cc65b3fdce9f5f51847d5e3b82"],"state_sha256":"612905420c310747686652ef5543cb40a0e97a1f48ac9b49dac0b07c31cf172a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GZIB9yE/S8EhvLn2+qYxDvdsOzPnSjn7fA/v+wsNa+ds03KwDHXf23qyeVCdBdQZWbFTu0UDOJcZQ9jXsKGYBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T16:52:09.588842Z","bundle_sha256":"fc5423e7b3007fd66ae5427e6687a2b346df6a45046d1b87dbbbf9377f82677b"}}