Author name cluster

Chao Feng

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

13 papers

2 author rows

AAAI Conference 2026 Conference Paper

MEML-GRPO: Heterogeneous Multi-Expert Mutual Learning for RLVR Advancement

Weitao Jia
Jinghui Lu
Haiyang Yu
Siqi Wang
Guozhi Tang
An-Lan Wang
Weijie Yin
Dingkang Yang

Recent advances demonstrate that reinforcement learning with verifiable rewards (RLVR) significantly enhances the reasoning capabilities of large language models (LLMs). However, standard RLVR faces challenges with reward sparsity, where zero rewards from consistently incorrect candidate answers provide no learning signal, particularly in challenging tasks. To address this,we propose Multi-Expert Mutual Learning GRPO (MEML-GRPO), an innovative framework that utilizes diverse expert prompts as system prompts to generate a broader range of responses, substantially increasing the likelihood of identifying correct solutions. Additionally, we introduce an inter-expert mutual learning mechanism that facilitates knowledge sharing and transfer among experts, further boosting the model’s performance through RLVR. Extensive experiments across multiple reasoning benchmarks show that MEML-GRPO delivers significant improvements, achieving an average performance gain of 4.89% with Qwen and 11.33% with Llama, effectively overcoming the core limitations of traditional RLVR methods.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Semi-Supervised Synthetic Data Generation with Fine-Grained Relevance Control for Short Video Search Relevance Modeling

Haoran Li
Zhiming Su
Junyan Yao
Enwei Zhang
Yang Ji
Yan Chen
Kan Zhou
Chao Feng

Synthetic data is widely adopted in embedding models to ensure diversity in training data distributions across dimensions such as difficulty, length, and language. However, existing prompt-based synthesis methods struggle to capture domain-specific data distributions, particularly in data-scarce domains, and often overlook fine-grained relevance diversity. In this paper, we present a Chinese short video dataset with 4-level relevance annotations, filling a critical resource void. Further, we propose a semi-supervised synthetic data pipeline where two collaboratively trained models generate domain-adaptive short video data with controllable relevance labels. Our method enhances relevance-level diversity by synthesizing samples for underrepresented intermediate relevance labels, resulting in a more balanced and semantically rich training data set. Extensive offline experiments show that the embedding model trained on our synthesized data outperforms those using data generated based on prompting or vanilla supervised fine-tuning(SFT). Moreover, we demonstrate that incorporating more diverse fine-grained relevance levels in training data enhances the model's sensitivity to subtle semantic distinctions, highlighting the value of fine-grained relevance supervision in embedding learning. In the search enhanced recommendation pipeline of Douyin's dual-column scenario, through online A/B testing, the proposed model increased click-through rate(CTR) by 1.45%, raised the proportion of Strong Relevance Ratio (SRR) by 4.9%, and improved the Image User Penetration Rate (IUPR) by 0.1054%.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement

Xiyao Wang
Zhengyuan Yang
Chao Feng
Hongjin Lu
Linjie Li
Chung-Ching Lin
Kevin Lin
Furong Huang

We introduce ThinkLite-VL, a family of visual reasoning models that achieve state-of-the-art (SoTA) performance using an order of magnitude fewer training samples, relying purely on reinforcement fine-tuning (RFT) self-improvement without any knowledge distillation. Our central insight is that sample difficulty critically influences RFT effectiveness: appropriately challenging examples can drive substantial reasoning improvements, even in low-data regimes. However, quantifying sample difficulty in a reliable and scalable manner remains non-trivial. To address this, we repurpose Monte Carlo Tree Search (MCTS) to measure sample difficulty via the number of reasoning iterations a vision-language model (VLM) requires to solve each instance. This MCTS-based selection procedure identifies samples that induce deeper reasoning while remaining solvable, allowing us to filter a high-quality subset from 70k open-source examples spanning math, natural image understanding, and chart comprehension. Using this approach, we select just 11k challenging samples for RFT on Qwen2. 5-VL-7B-Instruct and 7. 5k samples for Qwen2. 5-VL-72B-Instruct. The resulting models, ThinkLite-VL-7B and ThinkLite-VL-72B, significantly outperform their respective base models across eight visual reasoning benchmarks. In particular, ThinkLite-VL-7B improves the average performance of Qwen2. 5-VL-7B-Instruct by 7\% and surpasses all existing 7B-level models, as well as much larger models such as GPT-4o, O1 and Qwen2. 5-VL-72B, achieving a new SoTA score of 75. 1 on MathVista. ThinkLite-VL-72B further advances the SoTA frontier, achieving an accuracy of 79. 7 on MathVista and an average benchmark improvement of 4. 42 over the open-source SOTA. These results demonstrate that MCTS-guided difficulty filtering provides a scalable and effective path toward data-efficient self-improvement in multimodal reasoning.

PDF Details

ICRA Conference 2025 Conference Paper

This&That: Language-Gesture Controlled Video Generation for Robot Planning

Boyang Wang
Nikhil Sridhar
Chao Feng
Mark Van der Merwe
Adam Fishman
Nima Fazeli
Jeong Joon Park

Clear, interpretable instructions are invaluable for complex tasks, helping to clarify goals and anticipate necessary steps. In this work, we propose a robot learning framework for communicating, planning, and executing a wide range of tasks, dubbed This&That. This&That solves general tasks by leveraging video generative models, which, through training on internet-scale data, contain rich physical and semantic context. Through this work, we tackle three fundamental challenges in video-based planning: 1) unambiguous task communication with simple human instructions, 2) controllable video gen-eration that respects user intent, and 3) translating visual plans into robot actions. This& That adds gesture conditioning alongside language to generate video predictions as a suc-cinct and unambiguous alternative to existing language-only methods, especially in complex and uncertain environments. These video predictions are then fed into a behavior cloning architecture dubbed Diffusion Video to Action (DiVA), which outperforms prior state-of-the-art behavior cloning and video-based planning methods by substantial margins. Project web-site: https://this-and-that-vid.github.io/this-and-thatl.

Details

NeurIPS Conference 2025 Conference Paper

ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs

Xiyao Wang
Zhengyuan Yang
Chao Feng
Yuhang Zhou
Xiaoyu Liu
Yongyuan Liang
Ming Li
Ziyi Zang

Reinforcement learning (RL) has shown great effectiveness for fine-tuning large language models (LLMs) using tasks that are challenging yet easily verifiable, such as math reasoning or code generation. However, extending this success to visual perception in vision–language models (VLMs) has been impeded by the scarcity of vision-centric tasks that are simultaneously challenging and unambiguously verifiable. To this end, we introduce \textbf{ViCrit} (\textit{Visual Caption Hallucination Critic}), an RL proxy task that trains VLMs to localize a subtle, synthetic visual hallucination injected into paragraphs of human-written image captions. Starting from a 200-word captions, we inject a single, subtle visual description error—altering a few words on objects, attributes, counts, or spatial relations—and task the model to pinpoint the corrupted span given the image and the modified caption. This formulation preserves the full perceptual difficulty while providing a binary, exact-match reward that is easy to compute and unambiguous. Models trained with the \textbf{ViCrit Task} exhibit substantial gains across a variety of VL benchmarks. Crucially, the improvements transfer beyond natural-image training data to abstract image reasoning and visual math, showing promises of learning to perceive rather than barely memorizing seen objects. To facilitate evaluation, we further introduce \textbf{ViCrit-Bench}, a category-balanced diagnostic benchmark that systematically probes perception errors across diverse image domains and error types. Together, our results demonstrate that fine-grained hallucination criticism is an effective and generalizable objective for enhancing visual perception in VLMs.

PDF Details

TCS Journal 2023 Journal Article

Multi-objective evolutionary algorithms are generally good: Maximizing monotone submodular functions over sequences

Chao Qian
Dan-Xuan Liu
Chao Feng
Ke Tang

Evolutionary algorithms (EAs) are general-purpose optimization algorithms, inspired by natural evolution. Recent theoretical studies have shown that EAs can achieve good approximation guarantees for solving the problem classes of submodular optimization, which have a wide range of applications, such as maximum coverage, sparse regression, influence maximization, document summarization and sensor placement, just to name a few. Though they have provided some theoretical explanation for the general-purpose nature of EAs, the considered submodular objective functions are defined only over sets or multisets. To complement this line of research, this paper studies the problem class of maximizing monotone submodular functions over sequences, where the objective function depends on the order of items. We prove that for each kind of previously studied monotone submodular objective functions over sequences, i. e. , prefix monotone submodular functions, weakly monotone and strongly submodular functions, and DAG monotone submodular functions, a simple multi-objective EA, i. e. , GSEMO, can always reach or improve the best known approximation guarantee after running polynomial time in expectation. Note that these best-known approximation guarantees can be obtained only by different greedy-style algorithms before. Empirical studies on various applications, e. g. , accomplishing tasks, maximizing information gain, search-and-tracking and recommender systems, show the excellent performance of the GSEMO.

Details DOI

NeurIPS Conference 2022 Conference Paper

Recommender Forest for Efficient Retrieval

Chao Feng
Wuchao Li
Defu Lian
Zheng Liu
Enhong Chen

Recommender systems (RS) have to select the top-N items from a massive item set. For the sake of efficient recommendation, RS usually represents user and item as latent embeddings, and relies on approximate nearest neighbour search (ANNs) to retrieve the recommendation result. Despite the reduction of running time, the representation learning is independent of ANNs index construction; thus, the two operations can be incompatible, which results in potential loss of recommendation accuracy. To overcome the above problem, we propose the Recommender Forest (a. k. a. , RecForest), which jointly learns latent embedding and index for efficient and high-fidelity recommendation. RecForest consists of multiple k-ary trees, each of which is a partition of the item set via hierarchical balanced clustering such that each item is uniquely represented by a path from the root to a leaf. Given such a data structure, an encoder-decoder based routing network is developed: it first encodes the context, i. e. , user information, into hidden states; then, leveraging a transformer-based decoder, it identifies the top-N items via beam search. Compared with the existing methods, RecForest brings in the following advantages: 1) the false partition of the boundary items can be effectively alleviated by the use of multiple trees; 2) the routing operation becomes much more accurate thanks to the powerful transformer decoder; 3) the tree parameters are shared across different tree levels, making the index to be extremely memory-efficient. The experimental studies are performed on five popular recommendation datasets: with a significantly simplified training cost, RecForest outperforms competitive baseline approaches in terms of both recommendation accuracy and efficiency.

PDF Details

AAAI Conference 2021 Conference Paper

Multi-Objective Submodular Maximization by Regret Ratio Minimization with Theoretical Guarantee

Chao Feng
Chao Qian

Submodular maximization has attracted much attention due to its wide application and attractive property. Previous works mainly considered one single objective function, while there can be multiple ones in practice. As the objectives are usually conflicting, there exists a set of Pareto optimal solutions, attaining different optimal trade-offs among multiple objectives. In this paper, we consider the problem of minimizing the regret ratio in multi-objective submodular maximization, which is to find at most k solutions to approximate the whole Pareto set as well as possible. We propose a new algorithm RRMS by sampling representative weight vectors and solving the corresponding weighted sums of objective functions using some given α-approximation algorithm for single-objective submodular maximization. We prove that the regret ratio of the output of RRMS is upper bounded by 1−α+O( √ d − 1·( d k−d ) 1 d−1 ), where d is the number of objectives. This is the first theoretical guarantee for the situation with more than two objectives. When d = 2, it reaches the (1 − α + O(1/k))-guarantee of the only existing algorithm POLYTOPE. Empirical results on the applications of multiobjective weighted maximum coverage and Max-Cut show the superior performance of RRMS over POLYTOPE.

PDF Details

AAAI Conference 2020 Conference Paper

An Efficient Evolutionary Algorithm for Subset Selection with General Cost Constraints

Chao Bian
Chao Feng
Chao Qian
Yang Yu

In this paper, we study the problem of selecting a subset from a ground set to maximize a monotone objective function f such that a monotone cost function c is bounded by an upper limit. State-of-the-art algorithms include the generalized greedy algorithm and POMC. The former is an efﬁcient ﬁxed time algorithm, but the performance is limited by the greedy nature. The latter is an anytime algorithm that can ﬁnd better subsets using more time, but without any polynomial-time approximation guarantee. In this paper, we propose a new anytime algorithm EAMC, which employs a simple evolutionary algorithm to optimize a surrogate objective integrating f and c. We prove that EAMC achieves the best known approximation guarantee in polynomial expected running time. Experimental results on the applications of maximum coverage, inﬂuence maximization and sensor placement show the excellent performance of EAMC.

PDF Details

AAAI Conference 2020 Conference Paper

Subset Selection by Pareto Optimization with Recombination

Chao Qian
Chao Bian
Chao Feng

Subset selection, i. e. , to select a limited number of items optimizing some given objective function, is a fundamental problem with various applications such as unsupervised feature selection and sparse regression. By employing a multi-objective evolutionary algorithm (EA) with mutation only to optimize the given objective function and minimize the number of selected items simultaneously, the recently proposed POSS algorithm achieves state-of-the-art performance for subset selection. In this paper, we propose the PORSS algorithm by incorporating recombination, a characterizing feature of EAs, into POSS. We prove that PORSS can achieve the optimal polynomial-time approximation guarantee as POSS when the objective function is monotone, and can ﬁnd an optimal solution efﬁciently in some cases whereas POSS cannot. Extensive experiments on unsupervised feature selection and sparse regression show the superiority of PORSS over POSS. Our analysis also theoretically discloses that recombination from diverse solutions can be more likely than mutation alone to generate various variations, thereby leading to better exploration; this may be of independent interest for understanding the inﬂuence of recombination.

PDF Details

AAAI Conference 2019 Conference Paper

Unsupervised Feature Selection by Pareto Optimization

Chao Feng
Chao Qian
Ke Tang

Dimensionality reduction is often employed to deal with the data with a huge number of features, which can be generally divided into two categories: feature transformation and feature selection. Due to the interpretability, the efficiency during inference and the abundance of unlabeled data, unsupervised feature selection has attracted much attention. In this paper, we consider its natural formulation, column subset selection (CSS), which is to minimize the reconstruction error of a data matrix by selecting a subset of features. We propose an anytime randomized iterative approach POCSS, which minimizes the reconstruction error and the number of selected features simultaneously. Its approximation guarantee is well bounded. Empirical results exhibit the superior performance of POCSS over the state-of-the-art algorithms.

PDF Details

IJCAI Conference 2018 Conference Paper

Distributed Pareto Optimization for Subset Selection

Chao Qian
Guiying Li
Chao Feng
Ke Tang

The subset selection problem that selects a few items from a ground set arises in many applications such as maximum coverage, influence maximization, sparse regression, etc. The recently proposed POSS algorithm is a powerful approximation solver for this problem. However, POSS requires centralized access to the full ground set, and thus is impractical for large-scale real-world applications, where the ground set is too large to be stored on one single machine. In this paper, we propose a distributed version of POSS (DPOSS) with a bounded approximation guarantee. DPOSS can be easily implemented in the MapReduce framework. Our extensive experiments using Spark, on various real-world data sets with size ranging from thousands to millions, show that DPOSS can achieve competitive performance compared with the centralized POSS, and is almost always better than the state-of-the-art distributed greedy algorithm RandGreeDi.

PDF Details

IJCAI Conference 2018 Conference Paper

Sequence Selection by Pareto Optimization

Chao Qian
Chao Feng
Ke Tang

The problem of selecting a sequence of items from a universe that maximizes some given objective function arises in many real-world applications. In this paper, we propose an anytime randomized iterative approach POSeqSel, which maximizes the given objective function and minimizes the sequence length simultaneously. We prove that for any previously studied objective function, POSeqSel using a reasonable time can always reach or improve the best known approximation guarantee. Empirical results exhibit the superior performance of POSeqSel.

PDF Details