Publications & Projects
37 works across 4 research areas
LLM & Agent
Qwen3.8-Max: A New Bar for Coding and Cowork
Qwen Team, Alibaba Group
Qwen 3.8-Max scales to 2.4 trillion parameters with comprehensive improvements across coding, work, research, and long-horizon tasks.
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
Agents-A1 Team, Shanghai AI Lab
Open-source Agentic models specialized on Long-horizon tasks.
From Static Context to Calibrated Interactive RL: Mitigating Distribution Shift in Multi-turn Dialogue with Aligned Simulator
Xiaohua Wang*, Jiakang Yuan*, Zisu Huang, Muzhao Tian, Changze Lv, Kaitao Song, Tao Chen, Xiaoqing Zheng
Graph-search-based end-to-end MLE task solver.
Segment-Aligned Policy Optimization for Multi-Modal Reasoning
Lei Gao, Zhuoming Li, Mengxi Jia, Jiakang Yuan, Hongbo Sun, Hao Sun, Xuelong Li
A comprehensive benchmark specifically designed to evaluate the reasoning capability of MLLMs.
Reward Hacking in the Era of Large Models: Mechanisms, Emergent Misalignment, Challenges
Xiaohua Wang∗, Muzhao Tian∗, Yuqi Zeng∗, Zisu Huang∗, Jiakang Yuan∗, Bowen Chen∗, Jingwen Xu∗, Mingbo Zhou∗, Wenhao Liu, Muling Wu, Zhengkang Guo, Qi Qian, Yifei Wang, Feiran Zhang, Ruicheng Yin, Shihan Dou, Changze Lv, Tao Chen, Kaitao Song, Xu Tan, Tao Gui, Xiaoqing Zheng, Xuanjing Huang
System review of reward hacking.
Intern-S1-Pro: Scientific multimodal foundation model at trillion scale
Intern-S1 Team, Shanghai AI Lab
Intern-S1-Pro: one-trillion-parameter scientific multimodal foundation model.
InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
InternAgent Team, Shanghai AI Lab
InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery.
LSTM-MAS: A Long Short-Term Memory Inspired Multi-Agent System for Long-Context Understanding
Yichen Jiang, Jiakang Yuan, Chongjun Tu, Peng Ye, Tao Chen
A Multi-Agent System emulates LSTM’s hierarchical information flow and gated memory mechanisms for long-context understanding.
FlowSearch: Advancing Deep Research with Dynamic Structured Knowledge Flow
Yusong Hu, Runmin Ma, Yue Fan, Jinxin Shi, Zongsheng Cao, Yuhao Zhou, Jiakang Yuan, Shuaiyu Zhang, Shiyang Feng, Xiangchao Yan, Shufei Zhang, Wenlong Zhang, Lei Bai, Bo Zhang
A multi-agent system built upon the dynamic structured knowledge flow.
Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction
Muzhao Tian, Zisu Huang, Xiaohua Wang, Jingwen Xu, Zhengkang Guo, Qi Qian, Yuanzhe Shen, Kaitao Song, Jiakang Yuan, Changze Lv, Xiaoqing Zheng
A framework that allows users to dynamically regulate memory reliance.
AutoMLGen: Navigating Fine-Grained Optimization for Coding Agents
Shangheng Du, Xiangchao Yan, Dengyang Jiang, Jiakang Yuan, Yusong Hu, Xin Li, Liang He, Bo Zhang, Lei Bai
Graph-search-based end-to-end MLE task solver.
Chimera: Improving Generalist Model with Domain-Specific Experts
Tianshuo Peng*, Mingsheng Li*, Jiakang Yuan, Hongbin Zhou, Renqiu Xia, Renrui Zhang, Lei Bai, Song Mao, Bin Wang, Aojun Zhou, Botian Shi, Tao Chen, Bo Zhang, Xiangyu Yue
A scalable and low-cost multi-modal pipeline to boost existing LMMs with domain-specific experts.
MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs
Jiakang Yuan*, Tianshuo Peng*, Yilei Jiang, Yiting Lu, Renrui Zhang, Kaituo Feng, Chaoyou Fu, Tao Chen, Lei Bai, Bo Zhang, Xiangyu Yue
A comprehensive benchmark specifically designed to evaluate the reasoning capability of MLLMs.
InternAgent: When Agent Becomes the Scientist -- Building Closed-Loop System from Hypothesis to Verification
Bo Zhang*, Shiyang Feng*, Xiangchao Yan*, Jiakang Yuan*, Zhiyin Yu, Xiaohan He, Songtao Huang, Shaowei Hou, Zheng Nie, Zhilong Wang, Jinyao Liu, Runmin Ma, Tianshuo Peng, Peng Ye, Dongzhan Zhou, Shufei Zhang, Xiaosong Wang, Yilan Zhang, Meng Li, Zhongying Tu, Xiangyu Yue, Wangli Ouyang, Bowen Zhou, Lei Bai
A unified closed-loop multi-agent framework for Automatic Scientific Research.
SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing
Xiangchao Yan*, Shiyang Feng*, Jiakang Yuan, Renqiu Xia, Bin Wang, Bo Zhang, Lei Bai
SurveyForge automatically generates and refines the content of surveys.
Dolphin: Moving Towards Closed-loop Auto-research through Thinking, Practice, and Feedback
Jiakang Yuan, Xiangchao Yan, Shiyang Feng, Bo Zhang, Tao Chen, Botian Shi, Wanli Ouyang, Yu Qiao, Lei Bai, Bowen Zhou
A closed-loop auto-research framework that enables autonomous scientific research through thinking, practice, and feedback.
OmniCaptioner: One Captioner to Rule Them All
Yiting Lu*, Jiakang Yuan*, Zhen Li, Shitian Zhao, Qi Qin, Xinyue Li, Le Zhuo, Licheng Wen, Dongyang Liu, Yuewen Cao, Xiangchao Yan, Xin Li, Botian Shi, Tao Chen, Zhibo Chen, Lei Bai, Bo Zhang, Peng Gao
An all-in-one visual to text mapping tool.
GeoX: Geometric Problem Solving Through Unified Formalized Vision-Language Pre-training
Renqiu Xia*, Mingsheng Li*, Hancheng Ye, Wenjie Wu, Hongbin Zhou, Jiakang Yuan, Tianshuo Peng, Xinyu Cai, Xiangchao Yan, Bin Wang, Conghui He, Botian Shi, Tao Chen, Junchi Yan, Bo Zhang
A multi-modal large model focusing on geometric understanding and reasoning tasks with formalized visual-language pre-training.
AIGC
Lumina-Image 2.0: A Unified and Efficient Image Generative Framework
Qi Qin, Le Zhuo, Yi Xin, Ruoyi Du, Zhen Li, Bin Fu, Yiting Lu, Jiakang Yuan, Xinyue Li, Dongyang Liu, Xiangyang Zhu, Manyuan Zhang, Will Beddow, Erwann Millon, Victor Perez, Wenhai Wang, Conghui He, Bo Zhang, Xiaohong Liu, Hongsheng Li, Yu Qiao, Chang Xu, Peng Gao
State-of-the-art text-to-image generation model.
Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy
Hancheng Ye*, Jiakang Yuan*, Renqiu Xia, Xiangchao Yan, Tao Chen, Junchi Yan, Botian Shi, Bo Zhang
AdaptiveDiffusion adaptively reduces noise prediction steps during denoising guided by third-order latent difference.
3D Vision & Autonomous Driving
Trust Your Instincts: Confidence-Driven Test-Time RL for Vision-Language-Action Models
Siyao Chen, Jiakang Yuan, Jiaxin Wang, Tao Chen
Test-time reinforcement learning for vision-language-action models.
MLG-Stereo: ViT Based Stereo Matching with Multi-Stage Local-Global Enhancement
Haoyu Zhang, Jingyi Zhou, Peng Ye, Jiakang Yuan, Lin Zhang, Feng Xu, Tao Chen
Unlocking the full potential of ViT-based stereo matching.
Bi3D++: Hybrid Bi-domain Active Learning for Cross-domain 3D Object Detection
Jiakang Yuan, Xiangchao Yan, Botian Shi, Bo Zhang, Feng Xu, Yu Qiao, Tao Chen
A Bi-domain active learning approach which selects samples from both source and target domain to solve cross-domain 3D object detection.
SPOT: Scalable 3D Pre-training via Occupancy Prediction for Learning Transferable 3D Representations
Jiakang Yuan, Bo Zhang, Tao Chen, Botian Shi, Yu Qiao
Scalable 3D pre-training via occupancy prediction for learning transferable 3D representations.
Consistency-aware Self-Training for Iterative-based Stereo Matching
Jingyi Zhou*, Peng Ye*, Haoyu Zhang, Jiakang Yuan (Project Leader), Qiang Rao, YangChenXu Liu, Cailin Wu, Feng Xu, Tao Chen
CST-Stereo achieves impressive results in various scenarios including in-domain, domain adaptive and domain generalization.
All-in-One: Transferring Vision Foundation Models into Stereo Matching
Jingyi Zhou*, Haoyu Zhang*, Jiakang Yuan*, Peng Ye, Tao Chen, Hao Jiang, Meiya Chen, Yangyang Zhang
AIOStereo flexibly selects and transfers knowledge from multiple heterogeneous VFMs to a single stereo matching model. Rank 1st on Middlebury Stereo Evaluation.
3DET-Mamba: Causal Sequence Modelling for End-to-End 3D Object Detection
Mingsheng Li*, Jiakang Yuan*, Sijin Chen, Lin Zhang, Anyu Zhu, Xin Chen, Tao Chen
Exploit the potential of Mamba architecture on 3D scene-level perception for the first time.
Reg-TTA3D: Better Regression Makes Better Test-time Adaptive 3D Object Detection
Jiakang Yuan, Bo Zhang, Kaixiong Gong, Xiangyu Yue, Botian Shi, Yu Qiao, Tao Chen
A pseudo-label-based test-time adaptive 3D object detection method exploring a new task of test-time domain adaptive 3D detection.
ReSimAD: Zero-Shot 3D Domain Transfer for Autonomous Driving with Source Reconstruction and Target Simulation
Bo Zhang*, Xinyu Cai*, Jiakang Yuan, Donglin Yang, Jianfei Guo, Xiangchao Yan, Renqiu Xia, Botian Shi, Min Dou, Tao Chen, Si Liu, Junchi Yan, Yu Qiao
A Reconstruction-Simulation-Perception scheme for alleviating domain shifts in autonomous driving.
AD-PT: Autonomous Driving Pre-Training with Large-scale Point Cloud Dataset
Jiakang Yuan, Bo Zhang, Xiangchao Yan, Tao Chen, Botian Shi, Yikang Li, Yu Qiao
Build a large-scale pre-training point-cloud dataset with diverse data distribution, and learn generalizable representations.
Uni3D: A Unified Baseline for Multi-dataset 3D Object Detection
Bo Zhang, Jiakang Yuan, Botian Shi, Tao Chen, Yikang Li, Yu Qiao
A unified baseline to tackle multi-dataset 3D object detection from data-level and semantic-level.
Bi3D: Bi-domain Active Learning for Cross-domain 3D Object Detection
Jiakang Yuan, Bo Zhang, Xiangchao Yan, Tao Chen, Botian Shi, Yikang Li, Yu Qiao
A Bi-domain active learning approach which selects samples from both source and target domain to solve cross-domain 3D object detection.
Others
SciEvalKit: An Open-source Evaluation Toolkit for Scientific General Intelligence
SciEvalKit Team, Shanghai AI Lab
A unified benchmarking toolkit designed to evaluate AI models for science across a broad range of scientific disciplines and task capabilities.
Omniquality-R: Advanced Reward Models through All-Encompassing Quality Assessment
Yiting Lu, Fengbin Guan, Yixin Gao, Yan Zhong, Xinge Peng, Jiakang Yuan, Yihao Liu, Bo Zhang, Xin Li, Zhibo Chen, Weisi Lin
A unified reward modeling framework that transforms multi-task quality reasoning into continuous and interpretable reward signals .
Sequential Token Merging: Revisiting Hidden States
Yan Wen, Peng Ye, Lin Zhang, Baopu Li, Jiakang Yuan, Yaoxin Yang, Tao Chen
Novel sequential token reduction pipeline for Mamba.
Bidirectional Prototype-Reward co-Evolution for Test-Time Adaptation of Vision-Language Models
Xiaozhen Qiao, Peng Huang, Jiakang Yuan, Xianda Guo, Bowen Ye, Zhe Sun, Xuelong Li
Novel TTA framework for Vision-Language Models.
Learning Cross-Image Object Semantic Relation in Transformer for Few-Shot Fine-Grained Image Classification
Bo Zhang*, Jiakang Yuan*, Baopu Li, Tao Chen, Jiayuan Fan, Botian Shi
A Transformer-based double-helix model to achieve cross-image object semantic relation mining in a bidirectional and symmetrical manner.