arXiv:2502. 17666v5 Announce Type: replace-cross Abstract: Existing offline in-context reinforcement learning (ICRL) methods have predominantly relied on supervised training objectives, which are known to have limitations in offline RL settings.
arXiv:2502. 19544v3 Announce Type: replace Abstract: Leveraging offline data is a promising way to improve the sample efficiency of online reinforcement learning (RL).
arXiv:2509. 11259v2 Announce Type: replace-cross Abstract: Recent advancements in machine learning have largely been driven by foundation models (FMs) trained on large, diverse datasets, enabling them to generalize effectively to new, related tasks.
arXiv:2606. 25527v1 Announce Type: new Abstract: Online reinforcement learning (RL) agents increasingly depend on knowledge acquired offline to achieve practical efficiency.
arXiv:2607. 19450v1 Announce Type: cross Abstract: Large-scale online reinforcement learning (RL) is the predominant means of eliciting advanced abilities including long-term reasoning and agentic tool use in large language models (LLMs).
arXiv:2510. 01460v4 Announce Type: replace-cross Abstract: Offline-to-online reinforcement learning (RL) has emerged as a practical paradigm that leverages offline datasets for pretraining and online interactions for fine-tuning.
arXiv:2607. 27203v1 Announce Type: new Abstract: Pre-training followed by fine-tuning has become the dominant recipe for learning performant policies, and in value-based reinforcement learning (RL) this raises a natural question: given a pretrained policy, should the Q-function be pretrained on offline data too?
Pre-training followed by fine-tuning has become the dominant recipe for learning performant policies, and in value-based reinforcement learning (RL) this raises a natural question: given a pretrained policy, should the Q-function be pretrained on offline data too? Conventional wisdom suggests it should, but recent results show that online RL with a randomly-initialized Q-function can result in highly performant and reliable policies without needing to pretrain the Q-function.
arXiv:2306. 09712v2 Announce Type: replace Abstract: In reinforcement learning (RL), there are two major settings for interacting with the environment: online and offline.
Large-scale online reinforcement learning (RL) is the predominant means of eliciting advanced abilities including long-term reasoning and agentic tool use in large language models (LLMs). However, continuing to scale it across vast task domains of interest remains challenging in both computational infrastructure and cost, especially when considering RL as merely a one-off learning stage.
The paper introduces Q-Target Pretrained Transformers (QTPT), a method that replaces supervised behavior cloning with a Bellman-style Q‑target objective for in‑context reinforcement learning. QTPT retains the context‑conditioned Transformer architecture but learns to estimate action values using rewards and transitions from the context, rather than merely imitating offline actions. The authors provide theoretical analysis in stochastic linear bandits and finite‑horizon MDPs, demonstrating improved robustness to weak or suboptimal data, and empirically show gains over supervised pretraining on controlled RL benchmarks and extensions to D4RL Kitchen and AntMaze.
arXiv:2606. 10613v1 Announce Type: cross Abstract: Diffusion-based Q-learning has emerged as a powerful paradigm for offline reinforcement learning, but its reliance on multi-step denoising makes both training and inference computationally expensive and brittle.
arXiv:2605.10194v2 Announce Type: replace Abstract: On-policy self-distillation (OPSD) uses a model as its own teacher under privileged context, providing token-level supervision on the model's own r...