Related papers: GAC: Stabilizing Asynchronous RL Training for LLMs…

Reinforcement Learning (RL) is increasingly used in autonomous driving (AD) and shows clear advantages. However, most RL-based AD methods overlook policy structure design. An RL policy that only outputs short-timescale vehicle control…

Synchronous Reinforcement Learning (RL) post-training has emerged as a crucial step for enhancing Large Language Models (LLMs) with diverse capabilities. However, many systems designed to accelerate RL post-training still suffer from low…

Reinforcement learning is a general methodology of adaptive optimal control that has attracted much attention in various fields ranging from video game industry to robot manipulators. Despite its remarkable performance demonstrations, plain…

Comprehending natural language and following human instructions are critical capabilities for intelligent agents. However, the flexibility of linguistic instructions induces substantial ambiguity across language-conditioned tasks, severely…

This paper investigates the impact of using gradient norm reward signals in the context of Automatic Curriculum Learning (ACL) for deep reinforcement learning (DRL). We introduce a framework where the teacher model, utilizing the gradient…

Gradient dynamics play a central role in determining the stability and generalization of deep neural networks. In this work, we provide an empirical analysis of how variance and standard deviation of gradients evolve during training,…

Robotic foundation models achieve strong generalization by leveraging internet-scale vision-language representations, but their massive computational cost creates a fundamental bottleneck: high inference latency. In dynamic environments,…

Post-training large language models (LLMs) often suffers from catastrophic forgetting, where improvements on a target objective degrade previously acquired capabilities. Recent evidence suggests that this phenomenon is primarily driven by…

Recent advancements in domain generalization for deepfake detection have attracted significant attention, with previous methods often incorporating additional modules to prevent overfitting to domain-specific patterns. However, such…

Stochastic Gradient Descent (SGD) is the standard numerical method used to solve the core optimization problem for the vast majority of machine learning (ML) algorithms. In the context of large scale learning, as utilized by many Big Data…

Visual generation is dominated by three paradigms: AutoRegressive (AR), diffusion, and Visual AutoRegressive (VAR) models. Unlike AR and diffusion, VARs operate on heterogeneous input structures across their generation steps, which creates…

State-of-the-art model-based Reinforcement Learning (RL) approaches either use gradient-free, population-based methods for planning, learned policy networks, or a combination of policy networks and planning. Hybrid approaches that combine…

Dealing with high variance is a significant challenge in model-free reinforcement learning (RL). Existing methods are unreliable, exhibiting high variance in performance from run to run using different initializations/seeds. Focusing on…

Direct evaluation of LLMs on benchmarks can be misleading because comparatively strong performance may reflect task familiarity rather than capability. The train-before-test approach controls for task familiarity by giving each model…

Reinforcement learning post-training has substantially improved the reasoning accuracy of vision-language models, yet the resulting policies remain poorly calibrated. Terminal correctness rewards provide no gradient that penalizes confident…

Denoising-based generative models, particularly diffusion and flow matching algorithms, have achieved remarkable success. However, aligning their output distributions with complex downstream objectives, such as human preferences,…

Reinforcement learning (RL) is a promising data-driven approach for adaptive traffic signal control (ATSC) in complex urban traffic networks, and deep neural networks further enhance its learning power. However, centralized RL is infeasible…

This study presents a benchmark for evaluating action-constrained reinforcement learning (RL) algorithms. In action-constrained RL, each action taken by the learning system must comply with certain constraints. These constraints are crucial…

Regression is a fundamental task in machine learning that has garnered extensive attention over the past decades. The conventional approach for regression involves employing loss functions that primarily concentrate on aligning model…

Reinforcement learning (RL) for reachability specifications is fundamental in sequential decision-making, yet theoretical guarantees remain less explored. A recent work achieves asymptotic convergence to optimal policies. However, this…