# PAC-ACT: Post-training Actor-Critic for Action Chunking Transformers

arXiv: https://arxiv.org/abs/2607.09590 · HTML: https://mlsanity.com/paper/2607.09590/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 25 of its 32 references that are in the map.
- First posted on arXiv 10 Jul 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Offline Reinforcement Learning with Implicit Q-Learning](https://mlsanity.com/paper/2110.06169/) (2021)
- [What Matters in Learning from Offline Human Demonstrations for Robot Manipulation](https://mlsanity.com/paper/2108.03298/) (2021)
- [R3M: A Universal Visual Representation for Robot Manipulation](https://mlsanity.com/paper/2203.12601/) (2022)
- [robosuite: A Modular Simulation Framework and Benchmark for Robot Learning](https://mlsanity.com/paper/2009.12293/) (2020)
- [Implicit Behavioral Cloning](https://mlsanity.com/paper/2109.00137/) (2021)
- [Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning](https://mlsanity.com/paper/2108.06266/) (2021)
- [Behavior Transformers: Cloning k modes with one stone](https://mlsanity.com/paper/2206.11251/) (2022)
- [Diffusion Policy Policy Optimization](https://mlsanity.com/paper/2409.00588/) (2024)
- [DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution](https://mlsanity.com/paper/2411.02359/) (2024)
- [Reinforcement Learning with Action Chunking](https://mlsanity.com/paper/2507.07969/) (2025)
- [LeRobot: An Open-Source Library for End-to-End Robot Learning](https://mlsanity.com/paper/2602.22818/) (2026)
- [A Survey of Robot Manipulation in Contact](https://mlsanity.com/paper/2112.01942/) (2021)
- [Chunking the Critic: A Transformer-based Soft Actor-Critic with N-Step Returns](https://mlsanity.com/paper/2503.03660/) (2025)
- [Actor-Critic for Continuous Action Chunks: A Reinforcement Learning Framework for Long-Horizon Robotic Manipulation with Sparse Reward](https://mlsanity.com/paper/2508.11143/) (2025)
- [Coarse-to-fine Q-Network with Action Sequence for Data-Efficient Reinforcement Learning](https://mlsanity.com/paper/2411.12155/) (2024)

## Built on by (most cited first, 1 of 1)

- [The Latent That Never Was: A Forensic Re-run of the CVAE Ablation in Action Chunking Transformers](https://mlsanity.com/paper/2609.16745/) (2026)

## Abstract (arXiv)

Precision industrial contact manipulation requires reliable robot policies under pose perturbations and contact-force constraints. Vision-language-action models offer broad generalization but often introduce high inference latency and GPU-memory cost, while vision-action chunking policies are more suitable for real-time industrial control. However, these policies are usually trained by behavior cloning and suffer from distribution shift in contact-rich tasks. This paper proposes PAC-ACT, a reinforcement-learning post-training framework for pretrained Action Chunking Transformer policies. PAC-ACT reformulates policy optimization at the chunk level, constructs an ACT-transferred actor-critic architecture, and introduces a hybrid behavior-prior constraint to preserve the pretrained action distribution during online fine-tuning. Experiments on industrial precision-contact benchmarks show that PAC-ACT improves task success, contact stability, and force safety while retaining low latency and low GPU-memory usage. On the Contour task, PAC-ACT significantly reduces peak contact force and decreases the proportion of force readings above 60 N by 46 times. Sparse-reward ablations further show that the proposed behavior-prior constraint enables effective exploration under randomized initial poses.