{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2BDI7RYEWHJNSKPJ3AZAPGH5LG","short_pith_number":"pith:2BDI7RYE","schema_version":"1.0","canonical_sha256":"d0468fc704b1d2d929e9d8320798fd5992fa16a5b2a0d6ee7030045db322063f","source":{"kind":"arxiv","id":"2607.22531","version":1},"attestation_state":"computed","paper":{"title":"Twins: Learn to Predict Unified Representations with Focal Loss","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Bohao Li, Hao Wang, Jian-Wei Zhang, Kaixiong Gong, Liefeng Bo, Miles Yang, Mingzhe Zheng, Xiangyu Yue, Xin Cai, Yunlong Lin, Zhao Zhong","submitted_at":"2026-07-24T17:59:39Z","abstract_excerpt":"Unified multimodal models seek a shared visual token space that supports both multimodal understanding and image generation. Discrete methods unify the interface via a shared codebook, whereas continuous pipelines often rely on two disparate representations -- semantic features (e.g., ViT) for understanding and low-level latents (e.g., VAE) for synthesis -- resulting in mismatched latent spaces. We propose Twins, a unified continuous token space formed by channel-wise concatenating ViT and VAE features on the same token grid, so the sequence length is unchanged and attention cost does not incr"},"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":"2607.22531","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-24T17:59:39Z","cross_cats_sorted":[],"title_canon_sha256":"d9c9e01f27b07e15ec4fd4edd0f871eb1af0d6d326e4ed6bbb17ddf1ce008e69","abstract_canon_sha256":"b764161a519f894329ae452578aaa1674f0b6068d855bfe2dad9fdde01f1b822"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T01:21:32.432513Z","signature_b64":"kQhNdNWZR7xMRl85T/J9R+XFVvrCZVPr6D2SWdLI3HEAGgHqCJ7qCyhLBsJSOEJ9cYei0Qh5Ywml8GIHazo4CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0468fc704b1d2d929e9d8320798fd5992fa16a5b2a0d6ee7030045db322063f","last_reissued_at":"2026-07-27T01:21:32.431687Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T01:21:32.431687Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Twins: Learn to Predict Unified Representations with Focal Loss","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Bohao Li, Hao Wang, Jian-Wei Zhang, Kaixiong Gong, Liefeng Bo, Miles Yang, Mingzhe Zheng, Xiangyu Yue, Xin Cai, Yunlong Lin, Zhao Zhong","submitted_at":"2026-07-24T17:59:39Z","abstract_excerpt":"Unified multimodal models seek a shared visual token space that supports both multimodal understanding and image generation. Discrete methods unify the interface via a shared codebook, whereas continuous pipelines often rely on two disparate representations -- semantic features (e.g., ViT) for understanding and low-level latents (e.g., VAE) for synthesis -- resulting in mismatched latent spaces. We propose Twins, a unified continuous token space formed by channel-wise concatenating ViT and VAE features on the same token grid, so the sequence length is unchanged and attention cost does not incr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22531","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/2607.22531/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":"2607.22531","created_at":"2026-07-27T01:21:32.432108+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22531v1","created_at":"2026-07-27T01:21:32.432108+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22531","created_at":"2026-07-27T01:21:32.432108+00:00"},{"alias_kind":"pith_short_12","alias_value":"2BDI7RYEWHJN","created_at":"2026-07-27T01:21:32.432108+00:00"},{"alias_kind":"pith_short_16","alias_value":"2BDI7RYEWHJNSKPJ","created_at":"2026-07-27T01:21:32.432108+00:00"},{"alias_kind":"pith_short_8","alias_value":"2BDI7RYE","created_at":"2026-07-27T01:21:32.432108+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/2BDI7RYEWHJNSKPJ3AZAPGH5LG","json":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG.json","graph_json":"https://pith.science/api/pith-number/2BDI7RYEWHJNSKPJ3AZAPGH5LG/graph.json","events_json":"https://pith.science/api/pith-number/2BDI7RYEWHJNSKPJ3AZAPGH5LG/events.json","paper":"https://pith.science/paper/2BDI7RYE"},"agent_actions":{"view_html":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG","download_json":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG.json","view_paper":"https://pith.science/paper/2BDI7RYE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22531&json=true","fetch_graph":"https://pith.science/api/pith-number/2BDI7RYEWHJNSKPJ3AZAPGH5LG/graph.json","fetch_events":"https://pith.science/api/pith-number/2BDI7RYEWHJNSKPJ3AZAPGH5LG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG/action/storage_attestation","attest_author":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG/action/author_attestation","sign_citation":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG/action/citation_signature","submit_replication":"https://pith.science/pith/2BDI7RYEWHJNSKPJ3AZAPGH5LG/action/replication_record"}},"created_at":"2026-07-27T01:21:32.432108+00:00","updated_at":"2026-07-27T01:21:32.432108+00:00"}