aiqbench analysis
This paper applies Transformer sequence modeling to humanoid locomotion, using an auxiliary head that predicts short-horizon future proprioceptive states while keeping future information out of the action pathway. For ML readers, the useful idea is auxiliary future-state supervision as a way to give policies anticipatory structure. The reported gains are modest (93.96% vs. 90.49% for a vanilla Transformer) and come from one benchmark, so the abstract's "significantly" claim needs the full paper's variance and significance results.
ParkourFormer: Integrating Predictive Supervision and Sequence Modeling into Parkour Locomotion
arXiv:2605.25782v4 Announce Type: replace Abstract: Humanoid parkour requires locomotion policies to coordinate whole-body dynamics across rapidly changing terrains such as stairs, gaps, slopes, and obstacles. Existing learning-based humanoid locomotion policies typically infer actions from current and historical observations, while the future evolution of the robot state is represented only implicitly. Explicitly anticipating short-horizon state evolution is particularly beneficial for multiple terrains locomotion, where current actions must account for upcoming contact transitions and whole-body dynamics. We present ParkourForme, a Transformer-based sequence modeling framework that reformulates humanoid locomotion as a future-conditioned decision-making problem. The current robot state queries historical sensorimotor trajectories through cross-attention, while a lightweight prediction head forecasts short-horizon future proprioceptive states. The future states are predicted solely from historical observations and actions, with ground-truth future states used only as supervision targets, and are fused with temporal features to generate the current action without future-information leakage. We evaluate ParkourFormer on a diverse multi-terrain humanoid parkour benchmark including stairs, gaps, slopes, rough terrain, and obstacle traversal. ParkourFormer achieves a 93.96% mean traversal success rate, compared with 90.49% for a vanilla Transformer and 87.16% for a MoE-MLP policy, while maintaining a single policy across all terrain types. These results demonstrate that explicit future-state modeling significantly improves robustness and generalization for agile whole-body locomotion.
Abstract:Humanoid parkour requires locomotion policies to coordinate whole-body dynamics across rapidly changing terrains such as stairs, gaps, slopes, and obstacles. Existing learning-based humanoid locomotion policies typically infer actions from current and historical observations, while the future evolution of the robot state is represented only implicitly. Explicitly anticipating short-horizon state evolution is particularly beneficial for multiple terrains locomotion, where current actions must account for upcoming contact transitions and whole-body dynamics. We present ParkourForme, a Transformer-based sequence modeling framework that reformulates humanoid locomotion as a future-conditioned decision-making problem. The current robot state queries historical sensorimotor trajectories through cross-attention, while a lightweight prediction head forecasts short-horizon future proprioceptive states. The future states are predicted solely from historical observations and actions, with ground-truth future states used only as supervision targets, and are fused with temporal features to generate the current action without future-information leakage. We evaluate ParkourFormer on a diverse multi-terrain humanoid parkour benchmark including stairs, gaps, slopes, rough terrain, and obstacle traversal. ParkourFormer achieves a 93.96% mean traversal success rate, compared with 90.49% for a vanilla Transformer and 87.16% for a MoE-MLP policy, while maintaining a single policy across all terrain types. These results demonstrate that explicit future-state modeling significantly improves robustness and generalization for agile whole-body locomotion.
Robotics (cs.RO)
arXiv:2605.25782 [cs.RO]
(or arXiv:2605.25782v4 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2605.25782 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 296 words
Related coverage
ParkourFormer: Integrating Predictive Supervision and Sequence Modeling into Parkour Locomotion
arXiv · Jun 16, 2026
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · Oct 8, 2026
Mr.Dec: Daily-Scale Longitudinal Multimodal Modeling for 30-Day Readmission Prediction
arXiv · Aug 20, 2026
MT-ProtBERT: Multi-task Learning ProtBERT for Intrinsically Disordered Proteins Classification with Scarce Data
arXiv · Sep 23, 2026