From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models

Yixue Zhang; Wenke Xia; Kun Wu; Wenbo Yu; Pei Ren; Junnan Nie; Di Hu; Zhengping Che

cs.ROOct 29, 2026

From Reasoning to Action: Consequence-Grounded Alignment for Vision-Language-Action Models

Authors: Yixue Zhang, Wenke Xia, Kun Wu, Wenbo Yu, Pei Ren, Junnan Nie, Di Hu, Zhengping Che,

Organizations: Peking University, Beijing, China. · Beijing Innovation Center of Humanoid Robotics, Beijing, China. · Renmin University of China, Beijing, China. · Beijing Forestry University, Beijing, China.

Abstract

Reasoning informs decision-making by connecting intended actions with their expected consequences. Recent vision-language-action (VLA) models incorporate embodied chains of thought (ECoTs) to guide action generation. However, such guidance remains implicit, as actions are conditioned on ECoTs rather than explicitly optimized to align with them. We introduce CoRAct, a consequence-grounded reasoning-to-action alignment framework that explicitly aligns reasoning and action through reward-guided optimization in action noise space. CoRAct formulates two physically grounded ECoT components as verifiable signals: movement specifies intended motion, while waypoints encode expected spatial consequences. A unified Mixture-of-Transformers jointly models ECoT reasoning, action generation, and action-conditioned future prediction. For candidate actions generated from sampled noise, CoRAct computes alignment rewards based on movement consistency and agreement between predicted outcomes and specified waypoints. These rewards guide group-relative optimization of the initial noise distribution, steering action generation toward consistency with intended actions and expected consequences. Experiments across RoboCasa, VLABench, and real-world manipulation tasks demonstrate improvements in task success, generalization, and reasoning-execution alignment over strong ECoT-based VLA baselines. CoRAct also grounds externally provided ECoTs in executable behavior, enabling zero-shot adaptation to novel scenarios by leveraging knowledge from powerful vision-language models. Project page and dataset: https://onlyxingxingzyx.github.io/CoRAct/

Figures & tables

Fig. 1: Reasoning-to-action alignment paradigms. (a) Direct CoT-to-action learning leaves reasoning–execution correspondence implicit. (b) CoRAct explicitly bridges reasoning and execution through consequence-grounded optimization of the action noise distribution, aligning generated actions with articulated intent and expected outcomes.

Fig. 2: Framework of CoRAct. (a) CoRAct jointly models ECoT reasoning, action generation, and action-conditioned future prediction in a unified Mixture-of-Transformers, coupling actions with intent and consequence. (b) It then freezes the backbone and optimizes a conditional noise policy with movement and consequence rewards, steering action sampling toward better reasoning–execution alignment.

CoT Components		Alignment Objectives		Success Rate (%) $\uparrow$
Move. CoT	Waypt. CoT	Move. Align.	Waypt. Align.	VLABench	RoboCasa
				39.2%	68.6%
✓				44.3%	72.9%
✓	✓			47.5%	73.5%
✓	✓	✓		49.3%	74.6%
✓	✓	✓	✓	51.8%	75.5%

TABLE II: Ablation results of CoT components and alignment objectives.

Alignment		Consistency Metrics
Move. Align.	Waypt. Align.	Pos. cm $\downarrow$	Rot. deg $\downarrow$	Grip. % $\uparrow$	WP. px $\downarrow$
		0.401	0.534	69.789	5.998
✓		0.209	0.478	87.661	2.904
	✓	0.232	0.494	85.160	2.706
✓	✓	0.171	0.420	91.800	0.542

TABLE III: Ablation of movement and waypoint alignment on quantitative reasoning–execution consistency metrics.

Fig. 4: Illustration of our agent-guided correction pipeline, where GPT-5.6 Sol corrects the CoT generated by the VLM.

Components		Success Rate (%) $\uparrow$
R–A Align.	Agent Corr.	Insert Flower	Chem. Tube
		24.0	22.0
✓		26.0	30.0
	✓	52.0	40.0
✓	✓	66.0	52.0

TABLE IV: Closed-loop self-correction on representative VLABench tasks. We report task success rates (%).

Method	Tabletop organization	Image-guided word assembly	Language-guided fruit selection
$\pi_{0.5}$	33.3	26.6	56.6
$\mathrm{LaST}_{0}$	40.0	33.3	50.0
CoVer-VLA	33.3	26.6	50.0
CoRAct	56.6	46.6	63.3

TABLE V: Real-world evaluation on three long-horizon and semantic-intensive manipulation tasks.

VLMs for Robotics

Vision-language models for robotics connect visual and linguistic understanding to grounded navigation, manipulation, and robot control.

See all 664 papers

In-Context VLA: Endowing Vision-Language-Action Models with Language via In-Context Post-Training and Agentic Tool Use
Aug 6, 2026Jiarui Yang, Wen Huang, Jiale Zhang +2Reasoning for VLA Models ICL for Robotics
Nankai University · Tsinghua University · Swiss Federal Institute of Technology in Lausanne (EPFL)
ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation
Jun 16, 2026Tianyi Lu, Hui Zhang, Zijie Diao +8Long-Horizon Robotic Manipulation Language-Conditioned Robot Manipulation
Fudan University · Shanghai Innovation Institue · China Unicom +1
Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models
Jun 13, 2026Dianqiao Lei, Lianlei ShanEfficient Inference Efficient VLA Model Inference
1Tsinghua University, Beijing, China.

Aug 6, 2026J/K move · Enter open · S save

In-Context VLA: Endowing Vision-Language-Action Models with Language via In-Context Post-Training and Agentic Tool Use

Jiarui Yang, Wen Huang, Jiale Zhang, Maowei Hu+1

Nankai University · Tsinghua University · Swiss Federal Institute of Technology in Lausanne (EPFL)

Reasoning for VLA Models ICL for Robotics Continual Learning for VLA Models Robotic Manipulation Vision-Language-Action Models VLMs for Robotics+4

Open paper AskPDF arXiv

Vision-Language-Action (VLA) models have become the dominant recipe for generalist manipulation, yet they are almost universally trained by behavior cloning: a policy imitates expert action chunks conditioned on a static image and a fixed instruction. A natural remedy is to inject explicit reasoning through textual chain-of-thought (CoT). We show, both empirically and analytically, that free-form textual CoT degrades low-level control: the reasoning it produces is ungrounded, its latency breaks closed-loop timing, and, crucially, the reasoning and action tokens are optimized against conflicting objectives so that the policy learns to narrate rather than to act. We argue that what a VLA needs is not the ability to generate language, but the ability to consume grounded language. To this end we introduce \ourmethod{}, a framework that endows a VLA with language competence through (i) in-context post-training, in which perceptual evidence is injected as structured context and the model is supervised only on actions, and (ii) an agentic tool-use interface, in which the policy queries open-vocabulary detectors, monocular depth, and a vision--language model to actively acquire task-relevant information. Rather than emitting a single templated caption, our data engine produces diverse, paraphrased, and evidence-conditioned spatial descriptions, so that the policy learns to interpret language it has never seen verbatim. Across the RoboCasa-GR1, SimplerEnv, and LIBERO simulation benchmarks, together with 8 real-world robot manipulation tasks, our method consistently achieves SOTA results in both performance and efficiency when compared with CoT-based approaches under matched configurations.

Abstract

VLMs for Robotics

Explore similar work

In-Context VLA: Endowing Vision-Language-Action Models with Language via In-Context Post-Training and Agentic Tool Use

ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation

Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models