{"paper":{"title":"HDFlow: Hierarchical Diffusion-Flow Planning for Long-horizon Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"HDFlow uses a high-level diffusion planner to generate subgoals in latent space and a low-level rectified flow planner to produce trajectories for long-horizon robotic tasks.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chaoyi Xu, He Wang, Nandiraju Gireesh, Weiheng Liu, Yuanliang Ju, Yuxuan Wan","submitted_at":"2026-05-06T06:08:51Z","abstract_excerpt":"Recent advances in generative models have shown promise in generating behavior plans for long-horizon, sparse reward tasks. While these approaches have achieved promising results, they often lack a principled framework for hierarchical decomposition and struggle with the computational demands of real-time execution, due to their iterative denoising process. In this work, we introduce Hierarchical Diffusion-Flow (HDFlow), a novel hierarchical planning framework that optimally leverages the strengths of diffusion and rectified flow models to overcome the limitations of single-paradigm generative"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"HDFlow significantly outperforms state-of-the-art methods on four challenging furniture assembly tasks in both simulation and real-world, and generalizes to two long-horizon benchmarks comprising diverse locomotion and manipulation tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the learned latent space and subgoal conditioning allow the low-level rectified flow planner to generate feasible trajectories without additional feasibility checks or recovery mechanisms when high-level subgoals are unreachable.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"HDFlow pairs a high-level diffusion planner for subgoals with a low-level rectified flow planner for trajectories, outperforming prior methods on furniture assembly and locomotion-manipulation benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"HDFlow uses a high-level diffusion planner to generate subgoals in latent space and a low-level rectified flow planner to produce trajectories for long-horizon robotic tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6b35e9dc3eaccc0094734a19f848cc80f9819d0092e006717adc92fc4d94636f"},"source":{"id":"2605.04525","kind":"arxiv","version":2},"verdict":{"id":"bc05a89a-48b6-45ac-99d2-0d077d658323","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T16:22:31.381940Z","strongest_claim":"HDFlow significantly outperforms state-of-the-art methods on four challenging furniture assembly tasks in both simulation and real-world, and generalizes to two long-horizon benchmarks comprising diverse locomotion and manipulation tasks.","one_line_summary":"HDFlow pairs a high-level diffusion planner for subgoals with a low-level rectified flow planner for trajectories, outperforming prior methods on furniture assembly and locomotion-manipulation benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the learned latent space and subgoal conditioning allow the low-level rectified flow planner to generate feasible trajectories without additional feasibility checks or recovery mechanisms when high-level subgoals are unreachable.","pith_extraction_headline":"HDFlow uses a high-level diffusion planner to generate subgoals in latent space and a low-level rectified flow planner to produce trajectories for long-horizon robotic tasks."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.04525/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T23:01:19.753032Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:21:30.968655Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"b968a7a7b8a5c5518696a090d0c52b5753a7e353c76879dd75c3463dc8f13074"},"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"}