English
Related papers

Related papers: Extreme Region Policy Distillation

200 papers

This paper prescribes a suite of techniques for off-policy Reinforcement Learning (RL) that simplify the training process and reduce the sample complexity. First, we show that simple Deterministic Policy Gradient works remarkably well as…

Machine Learning · Computer Science 2020-06-30 Rasool Fakoor , Pratik Chaudhari , Alexander J. Smola

Reinforcement learning (RL) has been effective for post-training autoregressive (AR) language models, but extending these methods to diffusion language models (DLMs) is challenging due to intractable sequence-level likelihoods. Existing…

Recent work has shown that augmenting environments with language descriptions improves policy learning. However, for environments with complex language abstractions, learning how to ground language to observations is difficult due to…

Machine Learning · Computer Science 2022-10-04 Victor Zhong , Jesse Mu , Luke Zettlemoyer , Edward Grefenstette , Tim Rocktäschel

Distilling reasoning traces from strong large language models into smaller ones is a promising route to improve intelligence in resource-constrained settings. Existing approaches face a fundamental trade-off: offline distillation from…

Computation and Language · Computer Science 2026-05-15 Yumeng Zhang , Zhengbang Yang , Yevin Nikhel Goonatilake , Zhuangdi Zhu

Solving the Goal-Conditioned Reward Sparse (GCRS) task is a challenging reinforcement learning problem due to the sparsity of reward signals. In this work, we propose a new formulation of GCRS tasks from the perspective of the drifted…

Machine Learning · Computer Science 2020-05-01 Hao Sun , Xinyu Pan , Bo Dai , Dahua Lin , Bolei Zhou

Large language model post-training methods such as supervised fine-tuning (SFT), reinforcement learning (RL), and distillation are often analyzed through their loss functions: maximum likelihood, policy gradients, forward KL, reverse KL, or…

Machine Learning · Computer Science 2026-05-22 Dong Nie

Trust region methods, such as TRPO, are often used to stabilize policy optimization algorithms in reinforcement learning (RL). While current trust region strategies are effective for continuous control, they typically require a…

Artificial Intelligence · Computer Science 2018-02-26 Ofir Nachum , Mohammad Norouzi , Kelvin Xu , Dale Schuurmans

RLVR and OPD have become standard paradigms for post-training. We provide a unified analysis of these two paradigms in consolidating multiple expert capabilities into a single model, identifying capability loss in different ways: mixed RLVR…

Machine Learning · Computer Science 2026-05-01 Naibin Gu , Chenxu Yang , Qingyi Si , Chuanyu Qin , Dingyu Yao , Peng Fu , Zheng Lin , Weiping Wang , Nan Duan , Jiaqi Wang

Language model (LM) post-training (or alignment) involves maximizing a reward function that is derived from preference annotations. Direct Preference Optimization (DPO) is a popular offline alignment method that trains a policy directly on…

Machine Learning · Computer Science 2025-03-04 Adam Fisch , Jacob Eisenstein , Vicky Zayats , Alekh Agarwal , Ahmad Beirami , Chirag Nagpal , Pete Shaw , Jonathan Berant

Offline Reinforcement Learning (RL) aims to learn effective policies from a static dataset without requiring further agent-environment interactions. However, its practical adoption is often hindered by the need for explicit reward…

Machine Learning · Computer Science 2025-12-23 Gaurav Chaudhary , Laxmidhar Behera

Experience replay (ER) used in (deep) reinforcement learning is considered to be applicable only to off-policy algorithms. However, there have been some cases in which ER has been applied for on-policy algorithms, suggesting that…

Machine Learning · Computer Science 2024-09-16 Taisuke Kobayashi

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as an important paradigm for unlocking reasoning capabilities in large language models, exemplified by the success of OpenAI o1 and DeepSeek-R1. Currently, Group Relative…

Machine Learning · Computer Science 2026-01-08 Shijie Zhang , Kevin Zhang , Zheyuan Gu , Xiang Guo , Rujun Guo , Shaoyu Liu , Guanjun Jiang , Xiaozhao Wang

Enabling Large Language Models (LLMs) to continuously improve from environmental interactions is a central challenge in post-training. While on-policy self-distillation offers a promising paradigm, existing methods predominantly treat…

Machine Learning · Computer Science 2026-05-14 Yuwei Zhang , Sha Li , Changlong Yu , Qin Lu , Shuowei Jin , Chengyu Dong , Haoran Liu , Ilgee Hong , Xintong Li , Zhenyu Shi , Bing Yin , Jingbo Shang

Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models. While on-policy reinforcement learning can reduce forgetting, it…

Machine Learning · Computer Science 2026-01-28 Idan Shenfeld , Mehul Damani , Jonas Hübotter , Pulkit Agrawal

Deep reinforcement learning (RL) methods have significant potential for dialogue policy optimisation. However, they suffer from a poor performance in the early stages of learning. This is especially problematic for on-line learning with…

Computation and Language · Computer Science 2017-07-06 Pei-Hao Su , Pawel Budzianowski , Stefan Ultes , Milica Gasic , Steve Young

Model-free deep reinforcement learning (RL) methods have been successful in a wide variety of simulated domains. However, a major obstacle facing deep RL in the real world is their high sample complexity. Batch policy gradient methods offer…

Machine Learning · Computer Science 2017-03-01 Shixiang Gu , Timothy Lillicrap , Zoubin Ghahramani , Richard E. Turner , Sergey Levine

On-policy distillation (OPD) is a powerful paradigm for model alignment, yet its reliance on teacher logits restricts its application to white-box scenarios. We contend that structured semantic rubrics can serve as a scalable alternative to…

Machine Learning · Computer Science 2026-05-11 Junfeng Fang , Zhepei Hong , Mao Zheng , Mingyang Song , Gengsheng Li , Houcheng Jiang , Dan Zhang , Haiyun Guo , Xiang Wang , Tat-Seng Chua

Test-time reinforcement learning generates multiple candidate answers via repeated rollouts and performs online updates using pseudo-labels constructed by majority voting. To reduce overhead and improve exploration, prior work introduces…

Machine Learning · Computer Science 2026-05-28 Chu Zhao , Enneng Yang , Yuting Liu , Jianzhe Zhao , Guibing Guo

We study the problem of training and fine-tuning expressive policies with online reinforcement learning (RL) given an offline dataset. Training expressive policy classes with online RL present a unique challenge of stable value…

Machine Learning · Computer Science 2026-05-01 Perry Dong , Qiyang Li , Dorsa Sadigh , Chelsea Finn

Massive reinforcement learning (RL) data are typically collected to train policies offline without the need for interactions, but the large data volume can cause training inefficiencies. To tackle this issue, we formulate offline behavior…

Machine Learning · Computer Science 2024-10-31 Shiye Lei , Sen Zhang , Dacheng Tao