{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OHCIT4CCVZ4OCMLIRAAWRMRDBM","short_pith_number":"pith:OHCIT4CC","schema_version":"1.0","canonical_sha256":"71c489f042ae78e13168880168b2230b3027fbe2326e525db24ee02cac58ddf7","source":{"kind":"arxiv","id":"2605.22996","version":1},"attestation_state":"computed","paper":{"title":"CoMoGen: COntrollable MOtion Dynamics and Interactions with Mask-Guided Video GENeration","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adil Meric, Benjamin Busam, Christian Theobalt, Lin Geng Foo, Mert Kiray, Rishabh Dabral","submitted_at":"2026-05-21T19:51:20Z","abstract_excerpt":"We present CoMoGen, a controllable video generation framework that generates realistic interactive dynamics from a single binary mask sequence conditioned on an input image. CoMoGen introduces a lightweight MaskAdapter that encodes binary mask sequences into a latent residual signal, injected into the Multi Modal Diffusion Transformer (MMDiT) model through a cosine-weighted schedule. Unlike the hierarchical coarse-to-fine design of UNet architectures, MMDiT operates as a sequence of uniform transformer blocks, making it difficult to identify which layers are responsible for the motion generati"},"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":"2605.22996","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T19:51:20Z","cross_cats_sorted":[],"title_canon_sha256":"5129958c7a511f58e43df83842b4d2b732085df4f7a7e591df2408f1c6467106","abstract_canon_sha256":"cd1e4a7eda56bc1aed657ee76005caab7904bce178715061e0c36fe568a1a5fa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:33.719051Z","signature_b64":"yifrrfWnUVeDdQR4aUzELlsu8RclDK5XE/Un0CuSisB3TIX2hCtG71KgGcvsK+BOCA2vRxDxd/w528rlm61UAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71c489f042ae78e13168880168b2230b3027fbe2326e525db24ee02cac58ddf7","last_reissued_at":"2026-05-25T02:01:33.718346Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:33.718346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoMoGen: COntrollable MOtion Dynamics and Interactions with Mask-Guided Video GENeration","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adil Meric, Benjamin Busam, Christian Theobalt, Lin Geng Foo, Mert Kiray, Rishabh Dabral","submitted_at":"2026-05-21T19:51:20Z","abstract_excerpt":"We present CoMoGen, a controllable video generation framework that generates realistic interactive dynamics from a single binary mask sequence conditioned on an input image. CoMoGen introduces a lightweight MaskAdapter that encodes binary mask sequences into a latent residual signal, injected into the Multi Modal Diffusion Transformer (MMDiT) model through a cosine-weighted schedule. Unlike the hierarchical coarse-to-fine design of UNet architectures, MMDiT operates as a sequence of uniform transformer blocks, making it difficult to identify which layers are responsible for the motion generati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22996","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/2605.22996/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":"2605.22996","created_at":"2026-05-25T02:01:33.718437+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.22996v1","created_at":"2026-05-25T02:01:33.718437+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22996","created_at":"2026-05-25T02:01:33.718437+00:00"},{"alias_kind":"pith_short_12","alias_value":"OHCIT4CCVZ4O","created_at":"2026-05-25T02:01:33.718437+00:00"},{"alias_kind":"pith_short_16","alias_value":"OHCIT4CCVZ4OCMLI","created_at":"2026-05-25T02:01:33.718437+00:00"},{"alias_kind":"pith_short_8","alias_value":"OHCIT4CC","created_at":"2026-05-25T02:01:33.718437+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/OHCIT4CCVZ4OCMLIRAAWRMRDBM","json":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM.json","graph_json":"https://pith.science/api/pith-number/OHCIT4CCVZ4OCMLIRAAWRMRDBM/graph.json","events_json":"https://pith.science/api/pith-number/OHCIT4CCVZ4OCMLIRAAWRMRDBM/events.json","paper":"https://pith.science/paper/OHCIT4CC"},"agent_actions":{"view_html":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM","download_json":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM.json","view_paper":"https://pith.science/paper/OHCIT4CC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.22996&json=true","fetch_graph":"https://pith.science/api/pith-number/OHCIT4CCVZ4OCMLIRAAWRMRDBM/graph.json","fetch_events":"https://pith.science/api/pith-number/OHCIT4CCVZ4OCMLIRAAWRMRDBM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM/action/storage_attestation","attest_author":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM/action/author_attestation","sign_citation":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM/action/citation_signature","submit_replication":"https://pith.science/pith/OHCIT4CCVZ4OCMLIRAAWRMRDBM/action/replication_record"}},"created_at":"2026-05-25T02:01:33.718437+00:00","updated_at":"2026-05-25T02:01:33.718437+00:00"}