Paper Detail
Xirui Li, Zhe Liu, Xiaoqing Ye, Wenhua Han, Yifeng Pan, Junyu Han, Hengshuang Zhao
Multimodal driving planning faces a long-standing tension between two paradigms: scoring-based methods benefit from dense reward supervision but are confined to a fixed action vocabulary, while anchor-based methods generate proposals dynamically yet suffer from sparse supervision constrained to a single ground-truth trajectory. In this work, we propose FlowR2A, which resolves this tension by reframing simulation-based rewards from discriminative targets into generative conditions. By learning the reward-conditioned action distribution from dense trajectory-reward pairs with a flow-matching decoder, FlowR2A unifies the dense supervision of scoring-based methods with the proposal generation of anchor-based methods in a single generative model, forcing the model to internalize the correlation between an action and its outcomes in safety, progress, comfort, and rule compliance. To balance hard safety constraints against soft progress objectives, we introduce fine-grained per-timestep reward conditioning and reward noise augmentation. The generative formulation naturally supports controllable test-time sampling via reward guidance and anchored sampling, producing high-quality proposals. FlowR2A achieves state-of-the-art results on the NAVSIM v1 and v2 benchmarks, with multimodal proposals of substantially higher quality than prior methods.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{li2026flowr2a,
title = {FlowR2A: Learning Reward-to-Action Distribution for Multimodal Driving Planning},
author = {Xirui Li and Zhe Liu and Xiaoqing Ye and Wenhua Han and Yifeng Pan and Junyu Han and Hengshuang Zhao},
year = {2026},
abstract = {Multimodal driving planning faces a long-standing tension between two paradigms: scoring-based methods benefit from dense reward supervision but are confined to a fixed action vocabulary, while anchor-based methods generate proposals dynamically yet suffer from sparse supervision constrained to a single ground-truth trajectory. In this work, we propose FlowR2A, which resolves this tension by reframing simulation-based rewards from discriminative targets into generative conditions. By learning th},
url = {https://huggingface.co/papers/2606.24231},
keywords = {flow-matching decoder, reward-conditioned action distribution, multimodal driving planning, simulation-based rewards, discriminative targets, generative conditions, dense trajectory-reward pairs, anchor-based methods, scoring-based methods, reward guidance, anchored sampling, NAVSIM v1, NAVSIM v2, code available, huggingface daily},
eprint = {2606.24231},
archiveprefix = {arXiv},
}
{}