Recent Research Papers on Machine Learning

# Recent Research Papers on Machine Learning

## Summary of Work

1. **Hierarchical Rectified Flow Matching with Mini-Batch Couplings**: Introduces a hierarchical flow matching model to better capture multi-modality in velocity fields for generative modeling, with benefits shown in synthetic and imaging data.

2. **VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning**: Proposes VisionThink, a dynamic approach to visual token compression in vision-language models using reinforcement learning to optimize token use based on task complexity.

3. **Latent Policy Steering with Embodiment-Agnostic Pretrained World Models**: Focuses on reducing data collection for visuomotor policies by leveraging multi-embodiment datasets using optic flow and World Models, improving policy performance.

4. **Training Transformers with Enforced Lipschitz Constants**: Develops tools for maintaining norm-constrained weights in transformers to enhance stability and robustness, exploring performance tradeoffs.

5. **The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner**: Proposes TAIL to improve length generalization in language models via Turing Machine imitation learning, showing significant performance improvements on reasoning tasks.

## Papers

1. [Hierarchical Rectified Flow Matching with Mini-Batch Couplings](http://arxiv.org/pdf/2507.13350v1) - Yichi Zhang et al.

2. [VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning](http://arxiv.org/pdf/2507.13348v1) - Senqiao Yang et al.

3. [Latent Policy Steering with Embodiment-Agnostic Pretrained World Models](http://arxiv.org/pdf/2507.13340v1) - Yiqi Wang et al.

4. [Training Transformers with Enforced Lipschitz Constants](http://arxiv.org/pdf/2507.13338v1) - Laker Newhouse et al.

5. [The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner](http://arxiv.org/pdf/2507.13332v1) - Zhouqi Hua et al.