Related papers: CreFlow: Corrective Reflow for Sparse-Reward Embod…

Most reward models for visual generation reduce rich human judgments to a single unexplained score, discarding the reasoning that underlies preference. We show that teaching reward models to produce explicit, multi-dimensional critiques…

Image-based reinforcement learning (RL) faces significant challenges in generalization when the visual environment undergoes substantial changes between training and deployment. Under such circumstances, learned policies may not perform…

Instructional video editing applies edits to an input video using only text prompts, enabling intuitive natural-language control. Despite rapid progress, most methods still require fixed-length inputs and substantial compute. Meanwhile,…

We propose a new reinforcement learning (RL) formulation for training continuous-time score-based diffusion models for generative AI to generate samples that maximize reward functions while keeping the generated distributions close to the…

Video diffusion models lack explicit geometric supervision during training, leading to inconsistency artifacts such as object deformation, spatial drift, and depth violations in generated videos. To address this limitation, we propose a…

While online Reinforcement Learning has emerged as a crucial technique for aligning flow matching models with human preferences, current approaches are hindered by inefficient exploration during training rollouts. Relying on undirected…

Self-paced reinforcement learning (RL) aims to improve the data efficiency of learning by automatically creating sequences, namely curricula, of probability distributions over contexts. However, existing techniques for self-paced RL fail in…

Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising…

There is a surge of interest in using formal languages such as Linear Temporal Logic (LTL) to precisely and succinctly specify complex tasks and derive reward functions for Reinforcement Learning (RL). However, existing methods often assign…

Long-horizon precision manipulation in laboratory automation, such as pipette tip attachment and liquid transfer, requires policies that respect strict procedural logic while operating in continuous, high-dimensional state spaces. However,…

Post-training is the decisive step for converting a pretrained video generator into a production-oriented model that is instruction-following, controllable, and robust over long temporal horizons. This report presents a systematical…

Reinforcement learning (RL) has emerged as a powerful tool for tackling control problems, but its practical application is often hindered by the complexity arising from intricate reward functions with multiple terms. The reward hypothesis…

Instruction-driven image editing with unified multimodal generative models has advanced rapidly, yet their underlying visual reasoning remains limited, leading to suboptimal performance on reasoning-centric edits. Reinforcement learning…

Fine-tuning foundation models via reinforcement learning (RL) has proven promising for aligning to downstream objectives. In the case of diffusion models (DMs), though RL training improves alignment from early timesteps, critical issues…

Defining reward functions for skill learning has been a long-standing challenge in robotics. Recently, vision-language models (VLMs) have shown promise in defining reward signals for teaching robots manipulation skills. However, existing…

Designing dense rewards is crucial for reinforcement learning (RL), yet in robotics it often demands extensive manual effort and lacks scalability. One promising solution is to view task progress as a dense reward signal, as it quantifies…

Camera-controlled video generation has achieved remarkable progress in recent years. However, existing video-to-video re-rendering methods primarily rely on Supervised Fine-Tuning using synthetic datasets. At present, there is an extreme…

Visual model-based RL methods typically encode image observations into low-dimensional representations in a manner that does not eliminate redundant information. This leaves them susceptible to spurious variations -- changes in…

This paper presents CLIP-RLDrive, a new reinforcement learning (RL)-based framework for improving the decision-making of autonomous vehicles (AVs) in complex urban driving scenarios, particularly in unsignalized intersections. To achieve…

Reinforcement learning (RL) has become a central post-training tool for improving the reasoning abilities of large language models (LLMs). In these systems, the rollout, the trajectory sampled from a prompt to termination, including…