{"paper":{"title":"DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Discrete diffusion policies act as natural asynchronous executors because iterative unmasking makes inpainting native to them.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chenfeng Xu, Chensheng Peng, Chen Tang, Kaiwen Hong, Katherine Driggs-Campbell, Masayoshi Tomizuka, Pengcheng Wang","submitted_at":"2026-04-27T23:04:03Z","abstract_excerpt":"Unlike chatbots, physical AI must act while the world keeps evolving. Therefore, the inter-chunk pause of synchronous executors are fatal for dynamic tasks regardless of how fast the inference is. Asynchronous execution -- thinking while acting -- is therefore a structural requirement, and real-time chunking (RTC) makes it viable by recasting chunk transitions as inpainting: freezing committed actions and consistently generating the remainder. However, RTC with flow-matching policy is structurally suboptimal: its inpainting comes from inference-time corrections rather than the base policy, yie"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"discrete diffusion policies, which generate actions by iteratively unmasking, are natural asynchronous executors that resolve all limitations at once: they are fine-tuning free since inpainting is their native operation, while early stopping further provides adaptive guidance and reduces inference cost.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a discrete diffusion policy trained on standard offline data will produce consistent, high-quality inpainted continuations for committed action chunks in dynamic environments without quality degradation or the need for task-specific adaptations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Discrete diffusion policies support native asynchronous execution via unmasking for real-time chunking, delivering higher success rates and 0.7x inference cost versus flow-matching RTC on dynamic robotics benchmarks and real pick tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Discrete diffusion policies act as natural asynchronous executors because iterative unmasking makes inpainting native to them.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"70594df123f6d3278a9200435808d3ccf8f8a0f569c3095961b9fef00ebf4ba8"},"source":{"id":"2604.25050","kind":"arxiv","version":2},"verdict":{"id":"b1188067-bba4-492a-a2f3-41237c44c80a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T02:27:34.437853Z","strongest_claim":"discrete diffusion policies, which generate actions by iteratively unmasking, are natural asynchronous executors that resolve all limitations at once: they are fine-tuning free since inpainting is their native operation, while early stopping further provides adaptive guidance and reduces inference cost.","one_line_summary":"Discrete diffusion policies support native asynchronous execution via unmasking for real-time chunking, delivering higher success rates and 0.7x inference cost versus flow-matching RTC on dynamic robotics benchmarks and real pick tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a discrete diffusion policy trained on standard offline data will produce consistent, high-quality inpainted continuations for committed action chunks in dynamic environments without quality degradation or the need for task-specific adaptations.","pith_extraction_headline":"Discrete diffusion policies act as natural asynchronous executors because iterative unmasking makes inpainting native to them."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.25050/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T05:40:00.909962Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:32:12.770368Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"25c4035a507d082b62e9c6b8b32da2e1f377abbe81af4b0a580f74bf4927a88a"},"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"}