Wei Fu

NeurIPS Conference 2025 Conference Paper

AREAL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning

• Wei Fu
• Jiaxuan Gao
• Xujie Shen
• Chen Zhu
• Zhiyu Mei
• Chuyi He
• Shusheng Xu
• Guo Wei

Reinforcement learning (RL) has become a trending paradigm for training large language models (LLMs), particularly for reasoning tasks. Effective RL for LLMs requires massive parallelization and poses an urgent need for efficient training systems. Most existing large-scale RL systems for LLMs are synchronous by alternating generation and training in a batch setting, where the rollouts in each training batch are generated by the same (or latest) model. This stabilizes RL training but suffers from severe system-level inefficiency. Generation must wait until the longest output in the batch is completed before model update, resulting in GPU underutilization. We present AReaL, a fully asynchronous RL system that completely decouples generation from training. Rollout workers in AReaL continuously generate new outputs without waiting, while training workers update the model whenever a batch of data is collected. AReaL also incorporates a collection of system-level optimizations, leading to substantially higher GPU utilization. To stabilize RL training, AReaL balances the workload of rollout and training workers to control data staleness, and adopts a staleness-enhanced PPO variant to better handle outdated training samples. Extensive experiments on math and code reasoning benchmarks show that AReaL achieves up to 2. 77x training speedup compared to synchronous systems with the same number of GPUs and matched or even improved final performance. The code of AReaL is available at https: //github. com/inclusionAI/AReaL/.

NeurIPS Conference 2025 Conference Paper

How Far Are We from Optimal Reasoning Efficiency?

• Jiaxuan Gao
• Shu Yan
• Qixin Tan
• Lu Yang
• Shusheng Xu
• Wei Fu
• Zhiyu Mei
• Kaifeng Lyu

Large Reasoning Models (LRMs) demonstrate remarkable problem-solving capabilities through extended Chain-of-Thought (CoT) reasoning but often produce excessively verbose and redundant reasoning traces. This inefficiency incurs high inference costs and limits practical deployment. While existing fine-tuning methods aim to improve reasoning efficiency, assessing their efficiency gains remains challenging due to inconsistent evaluations. In this work, we introduce the reasoning efficiency frontiers, empirical upper bounds derived from fine-tuning a base LRM (DeepSeek-R1-Distill-Qwen-1. 5B/7B) across diverse approaches and training configurations. Based on these frontiers, we propose the Reasoning Efficiency Gap (REG), a unified metric quantifying deviations of any fine-tuned LRMs from these frontiers. Systematic evaluation on challenging mathematical benchmarks, AMC23, AIME24, and AIME25, reveals significant gaps in current methods: they either sacrifice accuracy for short length or use excessive tokens to achieve sub-optimal accuracies despite high overall accuracy. To reduce the efficiency gap, we propose REO-RL, a Reinforcement Learning algorithm that optimizes reasoning efficiency by targeting a sparse set of token budgets. Leveraging numerical integration over strategically selected budgets, REO-RL approximates the full efficiency objective with low error using a small set of token budgets. Experiments show that, compared to vanilla RL with outcome reward, REO-RL reduces the reasoning efficiency gap by 74. 5\% and 64. 2\% in the 1. 5B and 7B settings. The 7B LRM fine-tuned with REO-RL achieves reasoning conciseness surpassing frontier LRMs like Qwen3 and Claude Sonnet 3. 7. Ablation studies confirm the efficacy of our token budget strategy and highlight REO-RL’s flexibility across design choices. This work establishes a systematic framework for evaluating and optimizing reasoning efficiency in LRMs. We will release the related code, data, and models to support future research on efficient reasoning in LRMs.

EAAI Journal 2025 Journal Article

Self-labeled framework with semi-supervised ball K-means clustering-based synthetic example generation for semi-supervised classification in industrial applications

• Junnan Li
• Lufeng Wang
• Shun Fu
• Wei Fu
• Xin Pan

NeurIPS Conference 2024 Conference Paper

Hyper-opinion Evidential Deep Learning for Out-of-Distribution Detection

• Jingen Qu
• Yufei Chen
• Xiaodong Yue
• Wei Fu
• Qiguang Huang

Evidential Deep Learning (EDL), grounded in Evidence Theory and Subjective Logic (SL), provides a robust framework to estimate uncertainty for out-of-distribution (OOD) detection alongside traditional classification probabilities. However, the EDL framework is constrained by its focus on evidence that supports only single categories, neglecting the other collective evidences that could corroborate multiple in-distribution categories. This limitation leads to a diminished estimation of uncertainty and a subsequent decline in OOD detection performance. Additionally, EDL encounters the vanishing gradient problem within its fully-connected layers, further degrading classification accuracy. To address these issues, we introduce hyper-domain and propose Hyper-opinion Evidential Deep Learning (HEDL). HEDL extends the evidence modeling paradigm by explicitly integrating sharp evidence, which supports a singular category, with vague evidence that accommodates multiple potential categories. Additionally, we propose a novel opinion projection mechanism that translates hyper-opinion into multinomial-opinion, which is then optimized within the EDL framework to ensure precise classification and refined uncertainty estimation. HEDL integrates evidences across various categories to yield a holistic evidentiary foundation for achieving superior OOD detection. Furthermore, our proposed opinion projection method effectively mitigates the vanishing gradient issue, ensuring classification accuracy without additional model complexity. Extensive experiments over many datasets demonstrate our proposed method outperforms existing OOD detection methods.

ICML Conference 2024 Conference Paper

Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study

• Shusheng Xu
• Wei Fu
• Jiaxuan Gao
• Wenjie Ye
• Weilin Liu
• Zhiyu Mei
• Guangju Wang
• Chao Yu 0005

Reinforcement Learning from Human Feedback (RLHF) is currently the most widely used method to align large language models (LLMs) with human preferences. Existing RLHF methods can be roughly categorized as either reward-based or reward-free. Novel applications such as ChatGPT and Claude leverage reward-based methods that first learn a reward model and apply actor-critic algorithms, such as Proximal Policy Optimization (PPO). However, in academic benchmarks, state-of-the-art results are often achieved via reward-free methods, such as Direct Preference Optimization (DPO). Is DPO truly superior to PPO? Why does PPO perform poorly on these benchmarks? In this paper, we first conduct both theoretical and empirical studies on the algorithmic properties of DPO and show that DPO may have fundamental limitations. Moreover, we also comprehensively examine PPO and reveal the key factors for the best performances of PPO in fine-tuning LLMs. Finally, we benchmark DPO and PPO across a collection of RLHF testbeds, ranging from dialogue to code generation. Experiment results demonstrate that PPO is able to surpass other alignment methods in all cases and achieve state-of-the-art results in challenging code competitions.

ICRA Conference 2024 Conference Paper

Learning Agile Bipedal Motions on a Quadrupedal Robot

• Yunfei Li 0005
• Jinhan Li
• Wei Fu
• Yi Wu 0013

Can a quadrupedal robot perform bipedal motions like humans? Although developing human-like behaviors is more often studied on costly bipedal robot platforms, we present a solution over a lightweight quadrupedal robot that unlocks the agility of the quadruped in an upright standing pose and is capable of a variety of human-like motions. Our framework is with a hierarchical structure. At the low level is a motion-conditioned control policy that allows the quadrupedal robot to track desired base and front limb movements while balancing on two hind feet. The policy is commanded by a high-level motion generator that gives trajectories of parameterized human-like motions to the robot from multiple modalities of human input. We for the first time demonstrate various bipedal motions on a quadrupedal robot, and showcase interesting human-robot interaction modes including mimicking human videos, following natural language instructions, and physical interaction. The video is available at https://sites.google.com/view/bipedal-motions-quadruped.

IROS Conference 2024 Conference Paper

Robot Generating Data for Learning Generalizable Visual Robotic Manipulation

• Yunfei Li
• Ying Yuan
• Jingzhi Cui
• Haoran Huan
• Wei Fu
• Jiaxuan Gao
• Zekai Xu
• Yi Wu

It has been a popular trend in AI to pretrain foundation models on massive data. However, collecting sufficient offline training trajectories for robot learning is particularly expensive since valid control actions are required. Therefore, most existing robotic datasets are collected from human experts. We tackle such a data collection issue with a new framework called "robot self-teaching", which asks the robot to self-generate effective training data instead of relying on human demonstrators. Our key idea is to train a separate data-generation policy operating on the state space to automatically generate meaningful actions and trajectories with ever-growing complexities. Then, these generated data can be further used to train a visual policy with strong compositional generalization capabilities. We validate our framework in two visual manipulation testbeds, including a multi-object stacking domain and a popular RL benchmark "Franka kitchen". Experiments show that the final visual policy trained on self-generated data can accomplish novel testing goals that require long-horizon robot executions. Project website https://sites.google.com/view/robot-self-teaching.

ICLR Conference 2024 Conference Paper

SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores

• Zhiyu Mei
• Wei Fu
• Jiaxuan Gao
• Guangju Wang
• Huanchen Zhang
• Yi Wu 0013

The ever-growing complexity of reinforcement learning (RL) tasks demands a distributed system to efficiently generate and process a massive amount of data. However, existing open-source libraries suffer from various limitations, which impede their practical use in challenging scenarios where large-scale training is necessary. In this paper, we present a novel abstraction on the dataflows of RL training, which unifies diverse RL training applications into a general framework. Following this abstraction, we develop a scalable, efficient, and extensible distributed RL system called ReaLly Scalable RL (SRL), which allows efficient and massively parallelized training and easy development of customized algorithms. Our evaluation shows that SRL outperforms existing academic libraries, reaching at most 21x higher training throughput in a distributed setting. On learning performance, beyond performing and scaling well on common RL benchmarks with different RL algorithms, SRL can reproduce the same solution in the challenging hide-and-seek environment as reported by OpenAI with up to 5x speedup in wallclock time. Notably, SRL is the first in the academic community to perform RL experiments at a large scale with over 15k CPU cores. SRL anonymous repository is available at: https://anonymous.4open.science/r/srl-1E45/.

NeurIPS Conference 2023 Conference Paper

Iteratively Learn Diverse Strategies with State Distance Information

• Wei Fu
• Weihua Du
• Jingwei Li
• Sunli Chen
• Jingzhao Zhang
• Yi Wu

In complex reinforcement learning (RL) problems, policies with similar rewards may have substantially different behaviors. It remains a fundamental challenge to optimize rewards while also discovering as many diverse strategies as possible, which can be crucial in many practical applications. Our study examines two design choices for tackling this challenge, i. e. , diversity measure and computation framework. First, we find that with existing diversity measures, visually indistinguishable policies can still yield high diversity scores. To accurately capture the behavioral difference, we propose to incorporate the state-space distance information into the diversity measure. In addition, we examine two common computation frameworks for this problem, i. e. , population-based training (PBT) and iterative learning (ITR). We show that although PBT is the precise problem formulation, ITR can achieve comparable diversity scores with higher computation efficiency, leading to improved solution quality in practice. Based on our analysis, we further combine ITR with two tractable realizations of the state-distance-based diversity measures and develop a novel diversity-driven RL algorithm, State-based Intrinsic-reward Policy Optimization (SIPO), with provable convergence properties. We empirically examine SIPO across three domains from robot locomotion to multi-agent games. In all of our testing environments, SIPO consistently produces strategically diverse and human-interpretable policies that cannot be discovered by existing baselines.

ICLR Conference 2022 Conference Paper

Continuously Discovering Novel Strategies via Reward-Switching Policy Optimization

• Zihan Zhou 0002
• Wei Fu
• Bingliang Zhang
• Yi Wu 0013

We present Reward-Switching Policy Optimization (RSPO), a paradigm to discover diverse strategies in complex RL environments by iteratively finding novel policies that are both locally optimal and sufficiently different from existing ones. To encourage the learning policy to consistently converge towards a previously undiscovered local optimum, RSPO switches between extrinsic and intrinsic rewards via a trajectory-based novelty measurement during the optimization process. When a sampled trajectory is sufficiently distinct, RSPO performs standard policy optimization with extrinsic rewards. For trajectories with high likelihood under existing policies, RSPO utilizes an intrinsic diversity reward to promote exploration. Experiments show that RSPO is able to discover a wide spectrum of strategies in a variety of domains, ranging from single-agent navigation tasks and MuJoCo control to multi-agent stag-hunt games and the StarCraft II Multi-Agent Challenge.

ICML Conference 2022 Conference Paper

Revisiting Some Common Practices in Cooperative Multi-Agent Reinforcement Learning

• Wei Fu
• Chao Yu 0005
• Zelai Xu
• Jiaqi Yang
• Yi Wu 0013

Many advances in cooperative multi-agent reinforcement learning (MARL) are based on two common design principles: value decomposition and parameter sharing. A typical MARL algorithm of this fashion decomposes a centralized Q-function into local Q-networks with parameters shared across agents. Such an algorithmic paradigm enables centralized training and decentralized execution (CTDE) and leads to efficient learning in practice. Despite all the advantages, we revisit these two principles and show that in certain scenarios, e. g. , environments with a highly multi-modal reward landscape, value decomposition, and parameter sharing can be problematic and lead to undesired outcomes. In contrast, policy gradient (PG) methods with individual policies provably converge to an optimal solution in these cases, which partially supports some recent empirical observations that PG can be effective in many MARL testbeds. Inspired by our theoretical analysis, we present practical suggestions on implementing multi-agent PG algorithms for either high rewards or diverse emergent behaviors and empirically validate our findings on a variety of domains, ranging from the simplified matrix and grid-world games to complex benchmarks such as StarCraft Multi-Agent Challenge and Google Research Football. We hope our insights could benefit the community towards developing more general and more powerful MARL algorithms.