{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EEKS7JJPM3KOPKU67HDXQMGR6I","short_pith_number":"pith:EEKS7JJP","schema_version":"1.0","canonical_sha256":"21152fa52f66d4e7aa9ef9c77830d1f2304a924ad052ffaab205d84118a39421","source":{"kind":"arxiv","id":"2505.14705","version":1},"attestation_state":"computed","paper":{"title":"Beyond Modality Collapse: Representations Blending for Multimodal Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiawei Du, Joey Tianyi Zhou, Xin Zhang, Ziruo Zhang, Zuozhu Liu","submitted_at":"2025-05-16T03:00:56Z","abstract_excerpt":"Multimodal Dataset Distillation (MDD) seeks to condense large-scale image-text datasets into compact surrogates while retaining their effectiveness for cross-modal learning. Despite recent progress, existing MDD approaches often suffer from \\textit{\\textbf{Modality Collapse}}, characterized by over-concentrated intra-modal representations and enlarged distributional gap across modalities. In this paper, at the first time, we identify this issue as stemming from a fundamental conflict between the over-compression behavior inherent in dataset distillation and the cross-modal supervision imposed "},"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":"2505.14705","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-16T03:00:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"375d85f7de2a76dcfa0c7554b6aa0a33b1c056f929ebeba7533f53cef3d8bc69","abstract_canon_sha256":"23b2d3ba0ceb99a62e4964f174cdebadfd507c92e366e72a20935f19bd162d2f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:33.999864Z","signature_b64":"QO/Yg7bdEBTyq6r2lr2a5uoRLxDM8AnnN0hqHlhwh0eMeVuWaQJTKwARqdPR1PJBYrppBOE4rg9pvqWz3ra0Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21152fa52f66d4e7aa9ef9c77830d1f2304a924ad052ffaab205d84118a39421","last_reissued_at":"2026-07-05T11:06:33.999379Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:33.999379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Modality Collapse: Representations Blending for Multimodal Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiawei Du, Joey Tianyi Zhou, Xin Zhang, Ziruo Zhang, Zuozhu Liu","submitted_at":"2025-05-16T03:00:56Z","abstract_excerpt":"Multimodal Dataset Distillation (MDD) seeks to condense large-scale image-text datasets into compact surrogates while retaining their effectiveness for cross-modal learning. Despite recent progress, existing MDD approaches often suffer from \\textit{\\textbf{Modality Collapse}}, characterized by over-concentrated intra-modal representations and enlarged distributional gap across modalities. In this paper, at the first time, we identify this issue as stemming from a fundamental conflict between the over-compression behavior inherent in dataset distillation and the cross-modal supervision imposed "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14705","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/2505.14705/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":"2505.14705","created_at":"2026-07-05T11:06:33.999444+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14705v1","created_at":"2026-07-05T11:06:33.999444+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14705","created_at":"2026-07-05T11:06:33.999444+00:00"},{"alias_kind":"pith_short_12","alias_value":"EEKS7JJPM3KO","created_at":"2026-07-05T11:06:33.999444+00:00"},{"alias_kind":"pith_short_16","alias_value":"EEKS7JJPM3KOPKU6","created_at":"2026-07-05T11:06:33.999444+00:00"},{"alias_kind":"pith_short_8","alias_value":"EEKS7JJP","created_at":"2026-07-05T11:06:33.999444+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29464","citing_title":"Rank-Aware Hyperbolic Alignment for Vision-Language Dataset Distillation","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11705","citing_title":"CAST: Collapse-Aware multi-Scale Topology Fusion for Multimodal Coreset Selection","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I","json":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I.json","graph_json":"https://pith.science/api/pith-number/EEKS7JJPM3KOPKU67HDXQMGR6I/graph.json","events_json":"https://pith.science/api/pith-number/EEKS7JJPM3KOPKU67HDXQMGR6I/events.json","paper":"https://pith.science/paper/EEKS7JJP"},"agent_actions":{"view_html":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I","download_json":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I.json","view_paper":"https://pith.science/paper/EEKS7JJP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14705&json=true","fetch_graph":"https://pith.science/api/pith-number/EEKS7JJPM3KOPKU67HDXQMGR6I/graph.json","fetch_events":"https://pith.science/api/pith-number/EEKS7JJPM3KOPKU67HDXQMGR6I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I/action/storage_attestation","attest_author":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I/action/author_attestation","sign_citation":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I/action/citation_signature","submit_replication":"https://pith.science/pith/EEKS7JJPM3KOPKU67HDXQMGR6I/action/replication_record"}},"created_at":"2026-07-05T11:06:33.999444+00:00","updated_at":"2026-07-05T11:06:33.999444+00:00"}