papers

Publications (55)

cs.CV2025

Qwen3-VL Technical Report

Shuai Bai, Yuxuan Cai, Ruizhe Chen +61

We introduce Qwen3-VL, the most capable vision-language model in the Qwen series to date, achieving superior performance across a broad range of multimodal benchmarks. It natively…

cs.CL2021

InterBERT: Vision-and-Language Interaction for Multi-modal Pretraining

Junyang Lin, An Yang, Yichang Zhang +3

Multi-modal pretraining for learning high-level multi-modal representation is a further step towards deep learning and artificial intelligence. In this work, we propose a novel mod…

cs.CL2021

M6: A Chinese Multimodal Pretrainer

Junyang Lin, Rui Men, An Yang +22

In this work, we construct the largest dataset for multimodal pretraining in Chinese, which consists of over 1.9TB images and 292GB texts that cover a wide range of domains. We pro…

cs.CL2024

Language Models can Self-Lengthen to Generate Long Texts

Shanghaoran Quan, Tianyi Tang, Bowen Yu +7

Recent advancements in Large Language Models (LLMs) have significantly enhanced their ability to process long contexts, yet a notable gap remains in generating long, aligned output…

cs.CV2026

HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning

Shenzhi Wang, Shixuan Liu, Jing Zhou +8

Vision-language models (VLMs) show strong multimodal capabilities but still struggle with fine-grained vision-language reasoning. We find that long chain-of-thought (CoT) reasoning…

cs.CV2026

Qwen-Image-2.0 Technical Report

Bing Zhao, Chenfei Wu, Deqing Li +72

We present Qwen-Image-2.0, an omni-capable image generation foundation model that unifies high-fidelity generation and precise image editing within a single framework. Despite rece…

cs.CL2025

Rationales Are Not Silver Bullets: Measuring the Impact of Rationales on Model Performance and Reliability

Chiwei Zhu, Benfeng Xu, An Yang +4

Training language models with rationales augmentation has been shown to be beneficial in many existing works. In this paper, we identify that such a prevailing view does not hold c…

cs.CV2022

Transferring General Multimodal Pretrained Models to Text Recognition

Junyang Lin, Xuancheng Ren, Yichang Zhang +4

This paper proposes a new method, OFA-OCR, to transfer multimodal pretrained models to text recognition. Specifically, we recast text recognition as image captioning and directly t…

cs.RO2026

Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models

Haoqi Yuan, Zhixuan Liang, Anzhe Chen +20

Foundation models in language and multimodality achieve strong generalization by aligning heterogeneous data under a unified formulation and training at scale. In this report, we i…

cs.LG2025

Stabilizing Reinforcement Learning with LLMs: Formulation and Practices

Chujie Zheng, Kai Dang, Bowen Yu +10

This paper proposes a novel formulation for reinforcement learning (RL) with large language models, explaining why and under what conditions the true sequence-level reward can be o…

cs.CL2025

Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models

Yanzhao Zhang, Mingxin Li, Dingkun Long +9

In this work, we introduce the Qwen3 Embedding series, a significant advancement over its predecessor, the GTE-Qwen series, in text embedding and reranking capabilities, built upon…

cs.CL2024

Qwen2 Technical Report

An Yang, Baosong Yang, Binyuan Hui +59

This report introduces the Qwen2 series, the latest addition to our large language models and large multimodal models. We release a comprehensive suite of foundational and instruct…

cs.CL2024

Rethinking Data Selection at Scale: Random Selection is Almost All You Need

Tingyu Xia, Bowen Yu, Kai Dang +5

Supervised fine-tuning (SFT) is crucial for aligning Large Language Models (LLMs) with human instructions. The primary goal during SFT is to select a small yet representative subse…

cs.CL2024

Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement

An Yang, Beichen Zhang, Binyuan Hui +13

In this report, we present a series of math-specific large language models: Qwen2.5-Math and Qwen2.5-Math-Instruct-1.5B/7B/72B. The core innovation of the Qwen2.5 series lies in in…

cs.CL2022

Instance-wise Prompt Tuning for Pretrained Language Models

Yuezihan Jiang, Hao Yang, Junyang Lin +6

Prompt Learning has recently gained great popularity in bridging the gap between pretraining tasks and various downstream tasks. It freezes Pretrained Language Models (PLMs) and on…

cs.CL2018

SciDTB: Discourse Dependency TreeBank for Scientific Abstracts

An Yang, Sujian Li

Annotation corpus for discourse relations benefits NLP tasks such as machine translation and question answering. In this paper, we present SciDTB, a domain-specific discourse treeb…

cs.CL2018

Adaptations of ROUGE and BLEU to Better Evaluate Machine Reading Comprehension Task

An Yang, Kai Liu, Jing Liu +2

Current evaluation metrics to question answering based machine reading comprehension (MRC) systems generally focus on the lexical overlap between the candidate and reference answer…

cs.CL2026

Deeper is Not Always Better: Mitigating the Alignment Tax via Confident Layer Decoding

Xuanming Zhang, Sining Zhoubian, Yuxuan Chen +8

Autoregressive generation in large language models (LLMs) conventionally decodes from the final layer, assuming that deeper representations yield more reliable next-token predictio…

cs.CL2022

Prompt Tuning for Generative Multimodal Pretrained Models

Hao Yang, Junyang Lin, An Yang +3

Prompt tuning has become a new paradigm for model tuning and it has demonstrated success in natural language pretraining and even vision pretraining. In this work, we explore the t…

cs.CL2026

Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking

Mingxin Li, Yanzhao Zhang, Dingkun Long +9

In this report, we introduce the Qwen3-VL-Embedding and Qwen3-VL-Reranker model series, the latest extensions of the Qwen family built on the Qwen3-VL foundation model. Together, t…

cs.CL2025

CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratings

Shanghaoran Quan, Jiaxi Yang, Bowen Yu +14

With the increasing code reasoning capabilities of existing large language models (LLMs) and breakthroughs in reasoning models like OpenAI o1 and o3, there is a growing need to dev…

cs.CL2026

PLawBench: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice

Yuzhen Shi, Huanghai Liu, Yiran Hu +27

As large language models (LLMs) are increasingly applied to legal domain-specific tasks, evaluating their ability to perform legal work in real-world settings has become essential.…

cs.CL2023

Qwen Technical Report

Jinze Bai, Shuai Bai, Yunfei Chu +45

Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural language processing tasks that were previously thought to be exclusive to hu…

cs.CL2026

Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models

Boyi Deng, Xu Wang, Yaoning Wang +15

Large language models have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque, limiting our ability to inspec…

cs.LG2026

Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling

Yucheng Li, Huiqiang Jiang, Yang Xu +14

Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-To…

cs.CL2026

Qwen-AgentWorld: Language World Models for General Agents

Yuxin Zuo, Zikai Xiao, Li Sheng +30

A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigat…

cs.LG2021

M6-T: Exploring Sparse Expert Models and Beyond

An Yang, Junyang Lin, Rui Men +12

Mixture-of-Experts (MoE) models can achieve promising results with outrageous large amount of parameters but constant computation cost, and thus it has become a trend in model scal…

cs.CL2026

HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam

Weiqi Zhai, Zhihai Wang, Jinghang Wang +32

Humanity's Last Exam (HLE) has become a widely used benchmark for evaluating frontier large language models on challenging, multi-domain questions. However, community-led analyses…

cs.CL2025

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

Shenzhi Wang, Le Yu, Chang Gao +15

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful approach to enhancing the reasoning capabilities of Large Language Models (LLMs), while its mechanis…

cs.CL2026

Revealing Behavioral Plasticity in Large Language Models: A Token-Conditional Perspective

Liyuan Mao, Le Yu, Jing Zhou +7

In this work, we reveal that Large Language Models (LLMs) possess intrinsic behavioral plasticity-akin to chameleons adapting their coloration to environmental cues-that can be exp…

cs.CL2026

Are the Financial Reasoning from LLMs Credible? A Real World Test over Long-Horizon Statements

Xinke Tong, Xuanming Zhang, Tianyi Tang +10

Do Large Language Models (LLMs) possess genuine structural reasoning, or merely rely on surface-level pattern matching? The financial domain, demanding numerical precision and mult…

cs.CL2021

Learning Relation Alignment for Calibrated Cross-modal Retrieval

Shuhuai Ren, Junyang Lin, Guangxiang Zhao +5

Despite the achievements of large-scale multimodal pre-training approaches, cross-modal retrieval, e.g., image-text retrieval, remains a challenging task. To bridge the semantic ga…

cs.CL2025

RMTBench: Benchmarking LLMs Through Multi-Turn User-Centric Role-Playing

Hao Xiang, Tianyi Tang, Yang Su +10

Recent advancements in Large Language Models (LLMs) have shown outstanding potential for role-playing applications. Evaluating these capabilities is becoming crucial yet remains ch…

cs.LG2025

Group Sequence Policy Optimization

Chujie Zheng, Shixuan Liu, Mingze Li +9

This paper introduces Group Sequence Policy Optimization (GSPO), our stable, efficient, and performant reinforcement learning algorithm for training large language models. Unlike p…

cs.LG2026

Qwen-CUA: Native Computer Use for (almost) Everything

Dunjie Lu, Shuai Bai, Tianyi Bai +42

Native computer use offers a general interface for agents to operate almost any software available to people, but requires long-horizon state tracking, large-scale interactive expe…

cs.CL2025

WorldPM: Scaling Human Preference Modeling

Binghai Wang, Runji Lin, Keming Lu +17

Motivated by scaling laws in language modeling that demonstrate how test loss scales as a power law with model and dataset sizes, we find that similar laws exist in preference mode…

cs.CL2019

Machine Reading Comprehension: a Literature Review

Xin Zhang, An Yang, Sujian Li +1

Machine reading comprehension aims to teach machines to understand a text like a human and is a new challenging direction in Artificial Intelligence. This article summarizes recent…

cs.CL2025

Qwen3Guard Technical Report

Haiquan Zhao, Chenhan Yuan, Fei Huang +40

As large language models (LLMs) become more capable and widely used, ensuring the safety of their outputs is increasingly critical. Existing guardrail models, though useful in stat…

cs.MM2025

StyleSpeaker: Audio-Enhanced Fine-Grained Style Modeling for Speech-Driven 3D Facial Animation

An Yang, Chenyu Liu, Pengcheng Xia +1

Speech-driven 3D facial animation is challenging due to the diversity in speaking styles and the limited availability of 3D audio-visual data. Speech predominantly dictates the coa…

cs.LG2025

Soft Adaptive Policy Optimization

Chang Gao, Chujie Zheng, Xiong-Hui Chen +7

Reinforcement learning (RL) plays an increasingly important role in enhancing the reasoning capabilities of large language models (LLMs), yet stable and performant policy optimizat…

cs.CL2025

Qwen3 Technical Report

An Yang, Anfeng Li, Baosong Yang +57

In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency, a…

cs.DC2025

FastMPS: Revisit Data Parallel in Large-scale Matrix Product State Sampling

Yaojian Chen, Si-Qiu Gong, Lin Gan +5

Matrix Product State (MPS) is a versatile tensor network representation widely applied in quantum physics, quantum chemistry, and machine learning, etc. MPS sampling serves as a cr…

cs.CV2025

Binary-Gaussian: Compact and Progressive Representation for 3D Gaussian Segmentation

An Yang, Chenyu Liu, Jun Du +6

3D Gaussian Splatting (3D-GS) has emerged as an efficient 3D representation and a promising foundation for semantic tasks like segmentation. However, existing 3D-GS-based segmentat…

cs.CL2024

Qwen2.5-Coder Technical Report

Binyuan Hui, Jian Yang, Zeyu Cui +21

In this report, we introduce the Qwen2.5-Coder series, a significant upgrade from its predecessor, CodeQwen1.5. This series includes six models: Qwen2.5-Coder-(0.5B/1.5B/3B/7B/14B/…

cs.CL2025

ExpertPrompting: Instructing Large Language Models to be Distinguished Experts

Benfeng Xu, An Yang, Junyang Lin +4

The answering quality of an aligned large language model (LLM) can be drastically improved if treated with proper crafting of prompts. In this paper, we propose ExpertPrompting to…

cs.DS2025

GenTT: Generate Vectorized Codes for General Tensor Permutation

Yaojian Chen, Tianyu Ma, An Yang +3

Tensor permutation is a fundamental operation widely applied in AI, tensor networks, and related fields. However, it is extremely complex, and different shapes and permutation maps…

cs.CL2025

Qwen2.5 Technical Report

Qwen, :, An Yang +41

In this report, we introduce Qwen2.5, a comprehensive series of large language models (LLMs) designed to meet diverse needs. Compared to previous iterations, Qwen 2.5 has been sign…

cs.CV2022

OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework

Peng Wang, An Yang, Rui Men +7

In this work, we pursue a unified paradigm for multimodal pretraining to break the scaffolds of complex task/modality-specific customization. We propose OFA, a Task-Agnostic and Mo…

cs.RO2026

Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System

Jiazhao Zhang, Gengze Zhou, Hale Yin +32

Agentic navigation systems require a base navigation model whose observation strategy can be externally reconfigured at inference time, because instruction following, object search…

cs.CV2023

Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese

An Yang, Junshu Pan, Junyang Lin +4

The tremendous success of CLIP (Radford et al., 2021) has promoted the research and application of contrastive learning for vision-language pretraining. In this work, we construct…

cs.CV2025

Qwen-Image Technical Report

Chenfei Wu, Jiahao Li, Jingren Zhou +36

We present Qwen-Image, an image generation foundation model in the Qwen series that achieves significant advances in complex text rendering and precise image editing. To address th…

cs.CL2021

Sketch and Refine: Towards Faithful and Informative Table-to-Text Generation

Peng Wang, Junyang Lin, An Yang +4

Table-to-text generation refers to generating a descriptive text from a key-value table. Traditional autoregressive methods, though can generate text with high fluency, suffer from…

cs.LG2021

M6-10T: A Sharing-Delinking Paradigm for Efficient Multi-Trillion Parameter Pretraining

Junyang Lin, An Yang, Jinze Bai +9

Recent expeditious developments in deep learning algorithms, distributed training, and even hardware design for large models have enabled training extreme-scale models, say GPT-3 a…

cs.CL2025

Qwen2.5-1M Technical Report

An Yang, Bowen Yu, Chengyuan Li +25

We introduce Qwen2.5-1M, a series of models that extend the context length to 1 million tokens. Compared to the previous 128K version, the Qwen2.5-1M series have significantly enha…

cs.CV2022

OFASys: A Multi-Modal Multi-Task Learning System for Building Generalist Models

Jinze Bai, Rui Men, Hao Yang +15

Generalist models, which are capable of performing diverse multi-modal tasks in a task-agnostic way within a single model, have been explored recently. Being, hopefully, an alterna…