# SegDAC: Visual Generalization in Reinforcement Learning via Dynamic Object Tokens

arXiv: https://arxiv.org/abs/2508.09325 · HTML: https://mlsanity.com/paper/2508.09325/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 32 of its 38 references that are in the map.
- First posted on arXiv 12 Aug 2025.

## 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, 30 of 32)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Curiosity-driven Exploration by Self-supervised Prediction](https://mlsanity.com/paper/1705.05363/) (2017)
- [Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2301.08243/) (2023)
- [Object-Centric Learning with Slot Attention](https://mlsanity.com/paper/2006.15055/) (2020)
- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels](https://mlsanity.com/paper/2004.13649/) (2020)
- [Reinforcement Learning with Augmented Data](https://mlsanity.com/paper/2004.14990/) (2020)
- [Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning](https://mlsanity.com/paper/2107.09645/) (2021)
- [Recognize Anything: A Strong Image Tagging Model](https://mlsanity.com/paper/2306.03514/) (2023)
- [Improving Sample Efficiency in Model-Free Reinforcement Learning from Images](https://mlsanity.com/paper/1910.01741/) (2019)
- [Personalize Segment Anything Model with One Shot](https://mlsanity.com/paper/2305.03048/) (2023)
- [ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI](https://mlsanity.com/paper/2410.00425/) (2024)
- [LLaVA-Mini: Efficient Image and Video Large Multimodal Models with One Vision Token](https://mlsanity.com/paper/2501.03895/) (2025)
- [Generalization in Reinforcement Learning by Soft Data Augmentation](https://mlsanity.com/paper/2011.13389/) (2020)
- [THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation](https://mlsanity.com/paper/2402.08191/) (2024)
- [Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data Augmentation](https://mlsanity.com/paper/2107.00644/) (2021)
- [Open-Set Image Tagging with Multi-Grained Text Supervision](https://mlsanity.com/paper/2310.15200/) (2023)
- [Pre-Trained Image Encoder for Generalizable Visual Reinforcement Learning](https://mlsanity.com/paper/2212.08860/) (2022)
- [TorchRL: A data-driven decision-making library for PyTorch](https://mlsanity.com/paper/2306.00577/) (2023)
- [Look where you look! Saliency-guided Q-networks for generalization in visual Reinforcement Learning](https://mlsanity.com/paper/2209.09203/) (2022)
- [RL-ViGen: A Reinforcement Learning Benchmark for Visual Generalization](https://mlsanity.com/paper/2307.10224/) (2023)
- [A Recipe for Unbounded Data Augmentation in Visual Reinforcement Learning](https://mlsanity.com/paper/2405.17416/) (2024)
- [Shadow: Leveraging Segmentation Masks for Cross-Embodiment Policy Transfer](https://mlsanity.com/paper/2503.00774/) (2025)
- [Generalizable Visual Reinforcement Learning with Segment Anything Model](https://mlsanity.com/paper/2312.17116/) (2023)
- [CTRL-O: Language-Controllable Object-Centric Visual Representation Learning](https://mlsanity.com/paper/2503.21747/) (2025)

## Abstract (arXiv)

Visual reinforcement learning policies trained on pixel observations often struggle to generalize when visual conditions change at test time. Object-centric representations are a promising alternative, but most approaches use fixed-size slot representations, require image reconstruction, or need auxiliary losses to learn object decompositions. As a result, it remains unclear how to learn RL policies directly from object-level inputs without these constraints. We propose SegDAC, a Segmentation-Driven Actor-Critic that operates on a variable-length set of object token embeddings. At each timestep, text-grounded segmentation produces object masks from which spatially aware token embeddings are extracted. A transformer-based actor-critic processes these dynamic tokens, using segment positional encoding to preserve spatial information across objects. We ablate these design choices and show that both segment positional encoding and variable-length processing are individually necessary for strong performance. We evaluate SegDAC on 8 ManiSkill3 manipulation tasks under 12 visual perturbation types across 3 difficulty levels. SegDAC improves over prior visual generalization methods by 15% on easy, 66% on medium, and 88% on the hardest settings. SegDAC matches the sample efficiency of the state-of-the-art visual RL methods while achieving improved generalization under visual changes. Project Page: https://segdac.github.io/