Junyi Li

NeurIPS Conference 2025 Conference Paper

Incentivizing Dual Process Thinking for Efficient Large Language Model Reasoning

• Xiaoxue Cheng
• Junyi Li
• Zhenduo Zhang
• Xinyu Tang
• Xin Zhao
• Xinyu Kong
• Zhiqiang Zhang

Large reasoning models (LRMs) have demonstrated strong performance on complex reasoning tasks, but often suffer from overthinking, generating redundant content regardless of task difficulty. Inspired by the dual process theory in cognitive science, we propose Adaptive Cognition Policy Optimization (ACPO), a reinforcement learning framework that enables LRMs to achieve efficient reasoning through adaptive cognitive allocation and dynamic system switch. ACPO incorporates two key components: (1) introducing system-aware reasoning tokens to explicitly represent the thinking modes thereby making the model's cognitive process transparent, and (2) integrating online difficulty estimation and token length budget to guide adaptive system switch and reasoning during reinforcement learning. To this end, we propose a two-stage training strategy. The first stage begins with supervised fine-tuning to cold start the model, enabling it to generate reasoning paths with explicit thinking modes. In the second stage, we apply ACPO to further enhance adaptive system switch for difficulty-aware reasoning. Experimental results demonstrate that ACPO effectively reduces redundant reasoning while adaptively adjusting cognitive allocation based on task complexity, achieving efficient hybrid reasoning.

NeurIPS Conference 2025 Conference Paper

Reasoning Models Hallucinate More: Factuality-Aware Reinforcement Learning for Large Reasoning Models

• Junyi Li
• Hwee Tou Ng

Large language models (LLMs) have significantly advanced in reasoning tasks through reinforcement learning (RL) optimization, achieving impressive capabilities across various challenging benchmarks. However, our empirical analysis reveals a critical drawback: reasoning-oriented RL fine-tuning significantly increases the prevalence of hallucinations. We theoretically analyze the RL training dynamics, identifying high-variance gradient, entropy-induced randomness, and susceptibility to spurious local optima as key factors leading to hallucinations. To address this drawback, we propose Factuality-aware Step-wise Policy Optimization (FSPO), an innovative RL fine-tuning algorithm incorporating explicit factuality verification at each reasoning step. FSPO leverages automated verification against given evidence to dynamically adjust token-level advantage values, incentivizing factual correctness throughout the reasoning process. Experiments across mathematical reasoning and hallucination benchmarks using Qwen2. 5 and Llama models demonstrate that FSPO effectively reduces hallucinations while enhancing reasoning accuracy, substantially improving both reliability and performance.

AAAI Conference 2025 Conference Paper

Rethinking Transformer-Based Blind-Spot Network for Self-Supervised Image Denoising

• Junyi Li
• Zhilu Zhang
• Wangmeng Zuo

Blind-spot networks (BSN) have been prevalent neural architectures in self-supervised image denoising (SSID). However, most existing BSNs are conducted with convolution layers. Although transformers have shown the potential to overcome the limitations of convolutions in many image restoration tasks, the attention mechanisms may violate the blind-spot requirement, thereby restricting their applicability in BSN. To this end, we propose to analyze and redesign the channel and spatial attentions to meet the blind-spot requirement. Specifically, channel self-attention may leak the blind-spot information in multi-scale architectures, since the downsampling shuffles the spatial feature into channel dimensions. To alleviate this problem, we divide the channel into several groups and perform channel attention separately. For spatial self-attention, we apply an elaborate mask to the attention matrix to restrict and mimic the receptive field of dilated convolution. Based on the redesigned channel and window attentions, we build a Transformer-based Blind-Spot Network (TBSN), which shows strong local fitting and global perspective abilities. Furthermore, we introduce a knowledge distillation strategy that distills TBSN into smaller denoisers to improve computational efficiency while maintaining performance. Extensive experiments on real-world image denoising datasets show that TBSN largely extends the receptive field and exhibits favorable performance against state-of-the-art SSID methods.

NeurIPS Conference 2025 Conference Paper

VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning

• Senqiao Yang
• Junyi Li
• Xin Lai
• Jinming Wu
• Wei Li
• Zejun Ma
• Bei Yu
• Hengshuang Zhao

Recent advancements in vision-language models (VLMs) have improved performance by increasing the number of visual tokens, which are often significantly longer than text tokens. However, we observe that most real-world scenarios do not require such an extensive number of visual tokens. While the performance drops significantly in a small subset of OCR-related tasks, models still perform accurately in most other general VQA tasks with only 1/4 resolution. Therefore, we propose to dynamically process distinct samples with different resolutions, and present a new paradigm for visual token compression, namely, VisionThink. It starts with a downsampled image and smartly decides whether it is sufficient for problem solving. Otherwise, the model could output a special token to request the higher-resolution image. Compared to existing Efficient VLM methods that compress tokens using fixed pruning ratios or thresholds, VisionThink autonomously decides whether to compress tokens case by case. As a result, it demonstrates strong fine-grained visual understanding capability on OCR-related tasks, and meanwhile saves substantial visual tokens on simpler tasks. We adopt reinforcement learning and propose the LLM-as-Judge strategy to successfully apply RL to general VQA tasks. Moreoever, we carefully design a reward function and penalty mechanism to achieve a stable and reasonable image resize call ratio. Extensive experiments demonstrate the superiority, efficiency, and effectiveness of our method. All our code and data are open-sourced.

NeurIPS Conference 2024 Conference Paper

Exploring Context Window of Large Language Models via Decomposed Positional Vectors

• Zican Dong
• Junyi Li
• Xin Men
• Wayne X. Zhao
• Bingning Wang
• Zhen Tian
• weipeng chen
• Ji-Rong Wen

Transformer-based large language models (LLMs) typically have a limited context window, resulting in significant performance degradation when processing text beyond the length of the context window. Extensive studies have been proposed to extend the context window and achieve length extrapolation of LLMs, but there is still a lack of in-depth interpretation of these approaches. In this study, we explore the positional information within and beyond the context window for deciphering the underlying mechanism of LLMs. By using a mean-based decomposition method, we disentangle positional vectors from hidden states of LLMs and analyze their formation and effect on attention. Furthermore, when texts exceed the context window, we analyze the change of positional vectors in two settings, i. e. , direct extrapolation and context window extension. Based on our findings, we design two training-free context window extension methods, positional vector replacement and attention window extension. Experimental results show that our methods can effectively extend the context window length.

TMLR Journal 2024 Journal Article

Hessian Free Efficient Single Loop Iterative Differentiation Methods for Bi-Level Optimization Problems

• Peiran Yu
• Junyi Li
• Heng Huang

Bilevel optimization problems have been actively studied in recent machine learning research due to their broad applications. In this work, we investigate single-loop methods with iterative differentiation (ITD) for nonconvex bilevel optimization problems. For deterministic bilevel problems, we propose an efficient single-loop ITD-type method (ES-ITDM). Our method employs historical updates to approximate the hypergradient. More importantly, based on ES-ITDM, we propose a new method that avoids computing Hessians. This Hessian-free method requires fewer backpropagations and thus has a lower computational cost. We analyze the convergence properties of the proposed methods in two aspects. We provide the convergence rates of the sequences generated by ES-ITD based on the Kurdyka-\L ojasiewicz (KL) property. We also show that the Hessian-free stochastic ES-ITDM has the best-known complexity while has cheaper computation. The empirical studies show that our Hessian-free stochastic variant is more efficient than existing Hessian-free methods and other state-of-the-art bilevel optimization approaches.

NeurIPS Conference 2024 Conference Paper

Provably Faster Algorithms for Bilevel Optimization via Without-Replacement Sampling

• Junyi Li
• Heng Huang

Bilevel Optimization has experienced significant advancements recently with the introduction of new efficient algorithms. Mirroring the success in single-level optimization, stochastic gradient-based algorithms are widely used in bilevel optimization. However, a common limitation in these algorithms is the presumption of independent sampling, which can lead to increased computational costs due to the unique hyper-gradient structure in bilevel problems. To address this challenge, we study the example-selection strategy for bilevel optimization in this work. More specifically, we introduce a without-replacement sampling based algorithm which achieves a faster convergence rate compared to its counterparts that rely on independent sampling. Beyond the standard bilevel optimization formulation, we extend our discussion to conditional bilevel optimization and also two special cases: minimax and compositional optimization. Finally, we validate our algorithms over both synthetic and real-world applications. Numerical results clearly showcase the superiority of our algorithms.

JBHI Journal 2023 Journal Article

A Hierarchical Attention-Based Method for Sleep Staging Using Movement and Cardiopulmonary Signals

• Yujie Luo
• Junyi Li
• Kejing He
• William Cheuk

Sleep monitoring typically requires the uncomfortable and expensive polysomnography (PSG) test to determine the sleep stages. Body movement and cardiopulmonary signals provide an alternative way to perform sleep staging. In recent years, long-short term memory (LSTM) networks and convolutional neural networks (CNN) have dominated automatic sleep staging due to their better learning ability than machine learning classifiers. However, LSTM may lose information when dealing with long sequences, while CNN is not good at sequence modeling. As an improvement, we develop a hierarchical attention-based deep learning method for sleep staging using body movement, electrocardiogram (ECG), and abdominal breathing signals. We apply the multi-head self-attention to model the global context of feature sequences and coupled it with CNN to achieve a hierarchical self-attention weight assignment. We evaluate the performance of the method using two public datasets. Our method outperforms other baselines in the three sleep stages, achieving an accuracy of 84. 3 $\%$, an F1 score of 0. 8038, and a Cohen's Kappa coefficient of 0. 7036. The result demonstrates the effectiveness of the hierarchical self-attention mechanism when processing feature sequences in the sleep stage classification problem. This paper provides new possibilities for long-term sleep monitoring using movement and cardiopulmonary signals obtained from non-invasive devices.

NeurIPS Conference 2023 Conference Paper

Communication-Efficient Federated Bilevel Optimization with Global and Local Lower Level Problems

• Junyi Li
• Feihu Huang
• Heng Huang

Bilevel Optimization has witnessed notable progress recently with new emerging efficient algorithms. However, its application in the Federated Learning setting remains relatively underexplored, and the impact of Federated Learning's inherent challenges on the convergence of bilevel algorithms remain obscure. In this work, we investigate Federated Bilevel Optimization problems and propose a communication-efficient algorithm, named FedBiOAcc. The algorithm leverages an efficient estimation of the hyper-gradient in the distributed setting and utilizes the momentum-based variance-reduction acceleration. Remarkably, FedBiOAcc achieves a communication complexity $O(\epsilon^{-1})$, a sample complexity $O(\epsilon^{-1. 5})$ and the linear speed up with respect to the number of clients. We also analyze a special case of the Federated Bilevel Optimization problems, where lower level problems are locally managed by clients. We prove that FedBiOAcc-Local, a modified version of FedBiOAcc, converges at the same rate for this type of problems. Finally, we validate the proposed algorithms through two real-world tasks: Federated Data-cleaning and Federated Hyper-representation Learning. Empirical results show superior performance of our algorithms.

NeurIPS Conference 2023 Conference Paper

Federated Conditional Stochastic Optimization

• Xidong Wu
• Jianhui Sun
• Zhengmian Hu
• Junyi Li
• Aidong Zhang
• Heng Huang

Conditional stochastic optimization has found applications in a wide range of machine learning tasks, such as invariant learning, AUPRC maximization, and meta-learning. As the demand for training models with large-scale distributed data grows in these applications, there is an increasing need for communication-efficient distributed optimization algorithms, such as federated learning algorithms. This paper considers the nonconvex conditional stochastic optimization in federated learning and proposes the first federated conditional stochastic optimization algorithm (FCSG) with a conditional stochastic gradient estimator and a momentum-based algorithm (\emph{i. e. }, FCSG-M). To match the lower bound complexity in the single-machine setting, we design an accelerated algorithm (Acc-FCSG-M) via the variance reduction to achieve the best sample and communication complexity. Compared with the existing optimization analysis for Meta-Learning in FL, federated conditional stochastic optimization considers the sample of tasks. Extensive experimental results on various tasks validate the efficiency of these algorithms.

NeurIPS Conference 2023 Conference Paper

Resolving the Tug-of-War: A Separation of Communication and Learning in Federated Learning

• Junyi Li
• Heng Huang

Federated learning (FL) is a promising privacy-preserving machine learning paradigm over distributed data. In this paradigm, each client trains the parameter of a model locally and the server aggregates the parameter from clients periodically. Therefore, we perform the learning and communication over the same set of parameters. However, we find that learning and communication have fundamentally divergent requirements for parameter selection, akin to two opposite teams in a tug-of-war game. To mitigate this discrepancy, we introduce FedSep, a novel two-layer federated learning framework. FedSep consists of separated communication and learning layers for each client and the two layers are connected through decode/encode operations. In particular, the decoding operation is formulated as a minimization problem. We view FedSep as a federated bilevel optimization problem and propose an efficient algorithm to solve it. Theoretically, we demonstrate that its convergence matches that of the standard FL algorithms. The separation of communication and learning in FedSep offers innovative solutions to various challenging problems in FL, such as Communication-Efficient FL and Heterogeneous-Model FL. Empirical validation shows the superior performance of FedSep over various baselines in these tasks.

AAAI Conference 2022 Conference Paper

A Fully Single Loop Algorithm for Bilevel Optimization without Hessian Inverse

• Junyi Li
• Bin Gu
• Heng Huang

In this paper, we propose a novel Hessian inverse free Fully Single Loop Algorithm (FSLA) for bilevel optimization problems. Classic algorithms for bilevel optimization admit a double loop structure which is computationally expensive. Recently, several single loop algorithms have been proposed with optimizing the inner and outer variable alternatively. However, these algorithms not yet achieve fully single loop. As they overlook the loop needed to evaluate the hyper-gradient for a given inner and outer state. In order to develop a fully single loop algorithm, we first study the structure of the hypergradient and identify a general approximation formulation of hyper-gradient computation that encompasses several previous common approaches, e. g. back-propagation through time, conjugate gradient, etc. Based on this formulation, we introduce a new state variable to maintain the historical hyper-gradient information. Combining our new formulation with the alternative update of the inner and outer variables, we propose an efficient fully single loop algorithm. We theoretically show that the error generated by the new state can be bounded and our algorithm converges with the rate of O( −2 ). Finally, we verify the efficacy our algorithm empirically through multiple bilevel optimization based machine learning tasks. A long version of this paper can be found in: https: //arxiv. org/abs/2112. 04660.

IJCAI Conference 2022 Conference Paper

A Survey of Vision-Language Pre-Trained Models

• Yifan Du
• Zikang Liu
• Junyi Li
• Wayne Xin Zhao

As transformer evolves, pre-trained models have advanced at a breakneck pace in recent years. They have dominated the mainstream techniques in natural language processing (NLP) and computer vision (CV). How to adapt pre-training to the field of Vision-and-Language (V-L) learning and improve downstream task performance becomes a focus of multimodal learning. In this paper, we review the recent progress in Vision-Language Pre-Trained Models (VL-PTMs). As the core content, we first briefly introduce several ways to encode raw images and texts to single-modal embeddings before pre-training. Then, we dive into the mainstream architectures of VL-PTMs in modeling the interaction between text and image representations. We further present widely-used pre-training tasks, and then we introduce some common downstream tasks. We finally conclude this paper and present some promising research directions. Our survey aims to provide researchers with synthesis and pointer to related research.

NeurIPS Conference 2022 Conference Paper

Enhanced Bilevel Optimization via Bregman Distance

• Feihu Huang
• Junyi Li
• Shangqian Gao
• Heng Huang

Bilevel optimization has been recently used in many machine learning problems such as hyperparameter optimization, policy optimization, and meta learning. Although many bilevel optimization methods have been proposed, they still suffer from the high computational complexities and do not consider the more general bilevel problems with nonsmooth regularization. In the paper, thus, we propose a class of enhanced bilevel optimization methods with using Bregman distance to solve bilevel optimization problems, where the outer subproblem is nonconvex and possibly nonsmooth, and the inner subproblem is strongly convex. Specifically, we propose a bilevel optimization method based on Bregman distance (BiO-BreD) to solve deterministic bilevel problems, which achieves a lower computational complexity than the best known results. Meanwhile, we also propose a stochastic bilevel optimization method (SBiO-BreD) to solve stochastic bilevel problems based on stochastic approximated gradients and Bregman distance. Moreover, we further propose an accelerated version of SBiO-BreD method (ASBiO-BreD) using the variance-reduced technique, which can achieve a lower computational complexity than the best known computational complexities with respect to condition number $\kappa$ and target accuracy $\epsilon$ for finding an $\epsilon$-stationary point. We conduct data hyper-cleaning task and hyper-representation learning task to demonstrate that our new algorithms outperform related bilevel optimization approaches.

IJCAI Conference 2021 Conference Paper

Pretrained Language Model for Text Generation: A Survey

• Junyi Li
• Tianyi Tang
• Wayne Xin Zhao
• Ji-Rong Wen

Text generation has become one of the most important yet challenging tasks in natural language processing (NLP). The resurgence of deep learning has greatly advanced this field by neural generation models, especially the paradigm of pretrained language models (PLMs). In this paper, we present an overview of the major advances achieved in the topic of PLMs for text generation. As the preliminaries, we present the general task definition and briefly describe the mainstream architectures of PLMs for text generation. As the core content, we discuss how to adapt existing PLMs to model different input data and satisfy special properties in the generated text. We further summarize several important fine-tuning strategies for text generation. Finally, we present several future directions and conclude this paper. Our survey aims to provide text generation researchers a synthesis and pointer to related research.

NeurIPS Conference 2021 Conference Paper

SUPER-ADAM: Faster and Universal Framework of Adaptive Gradients

• Feihu Huang
• Junyi Li
• Heng Huang

Adaptive gradient methods have shown excellent performances for solving many machine learning problems. Although multiple adaptive gradient methods were recently studied, they mainly focus on either empirical or theoretical aspects and also only work for specific problems by using some specific adaptive learning rates. Thus, it is desired to design a universal framework for practical algorithms of adaptive gradients with theoretical guarantee to solve general problems. To fill this gap, we propose a faster and universal framework of adaptive gradients (i. e. , SUPER-ADAM) by introducing a universal adaptive matrix that includes most existing adaptive gradient forms. Moreover, our framework can flexibly integrate the momentum and variance reduced techniques. In particular, our novel framework provides the convergence analysis support for adaptive gradient methods under the nonconvex setting. In theoretical analysis, we prove that our SUPER-ADAM algorithm can achieve the best known gradient (i. e. , stochastic first-order oracle (SFO)) complexity of $\tilde{O}(\epsilon^{-3})$ for finding an $\epsilon$-stationary point of nonconvex optimization, which matches the lower bound for stochastic smooth nonconvex optimization. In numerical experiments, we employ various deep learning tasks to validate that our algorithm consistently outperforms the existing adaptive algorithms. Code is available at https: //github. com/LIJUNYI95/SuperAdam

AAAI Conference 2020 Conference Paper

Generating Realistic Stock Market Order Streams

• Junyi Li
• Xintong Wang
• Yaoyang Lin
• Arunesh Sinha
• Michael Wellman

We propose an approach to generate realistic and high- fidelity stock market data based on generative adversarial networks (GANs). Our Stock-GAN model employs a conditional Wasserstein GAN to capture history dependence of orders. The generator design includes specially crafted aspects including components that approximate the market’s auction mechanism, augmenting the order history with order-book constructions to improve the generation task. We perform an ablation study to verify the usefulness of aspects of our network structure. We provide a mathematical characterization of distribution learned by the generator. We also propose statistics to measure the quality of generated orders. We test our approach with synthetic and actual market data, compare to many baseline generative models, and find the generated data to be close to real data.

AIIM Journal 2017 Journal Article

Differential regulation analysis reveals dysfunctional regulatory mechanism involving transcription factors and microRNAs in gastric carcinogenesis

• Quanxue Li
• Junyi Li
• Wentao Dai
• Yi-Xue Li
• Yuan-Yuan Li

AAAI Conference 2016 Conference Paper

Estimating Text Intelligibility via Information Packaging Analysis

• Junyi Li

AAAI Conference 2015 Conference Paper

Fast and Accurate Prediction of Sentence Specificity

• Junyi Li
• Ani Nenkova

Recent studies have demonstrated that specificity is an important characterization of texts potentially beneficial for a range of applications such as multi-document news summarization and analysis of science journalism. The feasibility of automatically predicting sentence specificity from a rich set of features has also been confirmed in prior work. In this paper we present a practical system for predicting sentence specificity which exploits only features that require minimum processing and is trained in a semi-supervised manner. Our system outperforms the state-of-the-art method for predicting sentence specificity and does not require part of speech tagging or syntactic parsing as the prior methods did. With the tool that we developed — SPECITELLER — we study the role of specificity in sentence simplification. We show that specificity is a useful indicator for finding sentences that need to be simplified and a useful objective for simplification, descriptive of the differences between original and simplified sentences.