{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EGAC2JA3KZJRCXHRVQD6NYYKWE","short_pith_number":"pith:EGAC2JA3","schema_version":"1.0","canonical_sha256":"21802d241b5653115cf1ac07e6e30ab116412dd7335c112222779d37371db74a","source":{"kind":"arxiv","id":"2606.13289","version":1},"attestation_state":"computed","paper":{"title":"HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changlin Li, Guozhen Zhang, Jianbing Wu, Junzhe Li, Liefeng Bo, Limin Wang, Miles Yang, Tao Huang, Tianhui Song, Xiao Zhang, Xuerui Qiu, Yang Li, Yutao Cui, Zhao Zhong","submitted_at":"2026-06-11T12:46:07Z","abstract_excerpt":"Holistic visual tokenizers are fundamental to unified multimodal models (UMMs) as they map diverse visual inputs into a unified representation space. In this paper, we present HYDRA-X, the first UMM that unifies image and video tokenization within a single Vision Transformer (ViT). Our design is driven by two core challenges: efficiently injecting spatiotemporal reconstruction capability into a native ViT, and embedding image- and video-level semantic awareness into the latent space. To address the first, comprehensive ablations reveal two key findings: (1) frame-level causal temporal attentio"},"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":"2606.13289","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-11T12:46:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"feacc48b08e1a6d1d89781496986655d63e8cdc9ade8e72f91c8a313a942ea06","abstract_canon_sha256":"20550617386eac1419eb9e4c60bf0d164b3f354709a11d9fb8ab338e403be347"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:09:50.856126Z","signature_b64":"lc8rUhx4iTuDk1uGy9tQNhJZF2MMGP7HmnyjIjQe5/oHANX+8s7M5CjGeut8Z6WNFJqtSyNiCuWtJjT+2t8CAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21802d241b5653115cf1ac07e6e30ab116412dd7335c112222779d37371db74a","last_reissued_at":"2026-06-12T01:09:50.855468Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:09:50.855468Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changlin Li, Guozhen Zhang, Jianbing Wu, Junzhe Li, Liefeng Bo, Limin Wang, Miles Yang, Tao Huang, Tianhui Song, Xiao Zhang, Xuerui Qiu, Yang Li, Yutao Cui, Zhao Zhong","submitted_at":"2026-06-11T12:46:07Z","abstract_excerpt":"Holistic visual tokenizers are fundamental to unified multimodal models (UMMs) as they map diverse visual inputs into a unified representation space. In this paper, we present HYDRA-X, the first UMM that unifies image and video tokenization within a single Vision Transformer (ViT). Our design is driven by two core challenges: efficiently injecting spatiotemporal reconstruction capability into a native ViT, and embedding image- and video-level semantic awareness into the latent space. To address the first, comprehensive ablations reveal two key findings: (1) frame-level causal temporal attentio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13289","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/2606.13289/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":"2606.13289","created_at":"2026-06-12T01:09:50.855572+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.13289v1","created_at":"2026-06-12T01:09:50.855572+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13289","created_at":"2026-06-12T01:09:50.855572+00:00"},{"alias_kind":"pith_short_12","alias_value":"EGAC2JA3KZJR","created_at":"2026-06-12T01:09:50.855572+00:00"},{"alias_kind":"pith_short_16","alias_value":"EGAC2JA3KZJRCXHR","created_at":"2026-06-12T01:09:50.855572+00:00"},{"alias_kind":"pith_short_8","alias_value":"EGAC2JA3","created_at":"2026-06-12T01:09:50.855572+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/EGAC2JA3KZJRCXHRVQD6NYYKWE","json":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE.json","graph_json":"https://pith.science/api/pith-number/EGAC2JA3KZJRCXHRVQD6NYYKWE/graph.json","events_json":"https://pith.science/api/pith-number/EGAC2JA3KZJRCXHRVQD6NYYKWE/events.json","paper":"https://pith.science/paper/EGAC2JA3"},"agent_actions":{"view_html":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE","download_json":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE.json","view_paper":"https://pith.science/paper/EGAC2JA3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.13289&json=true","fetch_graph":"https://pith.science/api/pith-number/EGAC2JA3KZJRCXHRVQD6NYYKWE/graph.json","fetch_events":"https://pith.science/api/pith-number/EGAC2JA3KZJRCXHRVQD6NYYKWE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE/action/storage_attestation","attest_author":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE/action/author_attestation","sign_citation":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE/action/citation_signature","submit_replication":"https://pith.science/pith/EGAC2JA3KZJRCXHRVQD6NYYKWE/action/replication_record"}},"created_at":"2026-06-12T01:09:50.855572+00:00","updated_at":"2026-06-12T01:09:50.855572+00:00"}