Paper page - Flash-WAM: Modality-Aware Distillation for World Action Models

Source: https://huggingface.co/papers/2606.05254

Abstract

Flash-WAM introduces a modality-aware step-distillation framework for world-action models that achieves real-time inference by adapting consistency functions to different noise regimes in video and action streams.

World-action models(WAMs) jointly generate future video and robot actions through iterative diffusion, achieving strong performance on manipulation benchmarks but requiring tens of denoising steps, a cost that precludes real-time control.Step distillationhas emerged as the natural remedy, but off-the-shelf methods break down in the joint video-action setting because video and action streams use different SNR-shiftednoise schedulesand reach training with substantially differentmarginal noise distributions, an asymmetry that single-modality distillation methods cannot accommodate. We introduce Flash-WAM, amodality-awarestep-distillation framework inspired byconsistency distillationthat selects theconsistency functionfor each modality to match its noise regime: a linear-gradient-scaling parametrization for the action stream’s low-noise regime, paired with avariance-preserving parametrizationfor the video stream’s high-noise regime, grounded in a structural analysis of the consistency-function family that characterizes the achievablegradient scalingunder the consistency boundary condition. Instantiated on LingBot-VA, Flash-WAM compresses inference to a single step in each modality. OnRoboTwin 2.0, this reduces per-chunk latency from 8.1 seconds to 348 ms on NVIDIA L40S, a 23{times} speedup that enablesreal-time inference. Flash-WAM preserves task success on simulation benchmarks (85.5%RoboTwin 2.0, 95.7%LIBERO) and substantially recovers real-world performance (60% average on a Unitree G1 humanoid robot), while naiveconsistency distillationdrops to 24% at the same step budget.

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2606\.05254

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.05254 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.05254 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.05254 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.