mlsanityData to 14 Oct 2026

MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action

arXiv 2606.06245 ↗First posted 4 Jun 2026cs.RO cs.AIcited by 0

On the map: Vision-language-action models → Reasoning and chain-of-thought in VLA

Abstract (arXiv)

Vision-Language-Action (VLA) policies remain brittle in long-horizon control, where one-pass action decoding offers limited inference-time deliberation. Explicit chain-of-thought adds reasoning depth but incurs token-generation latency and an indirect text-to-action interface. We propose MPCoT, a reward-guided multi-path latent reasoning framework with configurable depth K and width M. MPCoT initializes M latent hypotheses, refines them for K weight-tied steps, and softly aggregates them before action decoding. A training-only path-preference objective combines expert-trajectory consistency, frozen Qwen3-VL progress scores, and endpoint-consistency feedback to supervise the path scorer. Under matched protocols on LIBERO and CALVIN, MPCoT improves long-horizon performance; ablations support the contributions of depth, width, soft aggregation, and reward supervision. On five real-world ALOHA Mini bimanual tasks, average success increases from 71.3% to 82.0% over the matched OpenVLA-OFT baseline. These results support latent deliberation as a means of improving execution while preserving the action interface and generating no reasoning tokens.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning and chain-of-thought in VLA, one of 13 directions in Vision-language-action models. Builds on 31 of its 31 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

31 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models27 Mar

Other references · 30

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  3. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  4. 2024π0: A Vision-Language-Action Flow Model for General Robot Controlcited by 2,407
  5. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,348
  6. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,071
  7. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  8. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,410
  9. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 883
  10. 2024Training Large Language Models to Reason in a Continuous Latent Spacecited by 797
  11. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 665
  12. 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 519
18 more
  1. 2025UniVLA: Learning to Act Anywhere with Task-centric Latent Actionscited by 410
  2. 2025WorldVLA: Towards Autoregressive Action World Modelcited by 367
  3. 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 304
  4. 2025CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillationcited by 225
  5. 2024Compressed Chain of Thought: Efficient Reasoning Through Dense Representationscited by 201
  6. 2024Predictive Inverse Dynamics Models are Scalable Learners for Robotic Manipulationcited by 161
  7. 2025VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Modelcited by 145
  8. 2025Interactive Post-Training for Vision-Language-Action Modelscited by 135
  9. 2025Unified Vision-Language-Action Modelcited by 134
  10. 2025NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Taskscited by 127
  11. 2025SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMscited by 116
  12. 2025Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoningcited by 91
  13. 2025PD-VLA: Accelerating Vision-Language-Action Model Integrated with Action Chunking via Parallel Decodingcited by 88
  14. 2025Efficient Reasoning with Hidden Thinkingcited by 75
  15. 2025FLOWER: Democratizing Generalist Robot Policies with Efficient Vision-Language-Action Flow Policiescited by 57
  16. 2025AVA-VLA: Improving Vision-Language-Action models with Active Visual Attentioncited by 25
  17. 2026Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoningcited by 18
  18. 2025Controlling Thinking Speed in Reasoning Modelscited by 10

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Recompose and Refine Latent Reasoning Flows for Vision-Language-Action Models
  2. 2026Continuous Reasoning for Vision-Language-Action
  3. 2026Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
  4. 2026Explicit Language Memory for Long-Horizon Planning in Vision-Language-Action Models
  5. 2026Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models
  6. 2025VLA-R1: Enhancing Reasoning in Vision-Language-Action Models
  7. 2026Latent Reasoning VLA: Latent Thinking and Prediction for Vision-Language-Action Models
  8. 2025DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models