Licheng Wen

2026

1. ACL Findings

   The Agent’s First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios

   Daocheng Fu*, Jianbiao Mei*, Rong Wu*, and 7 more authors

   Findings of the Association for Computational Linguistics (ACL), 2026

   Abs arXiv Bib Code

   The rapid evolution of Multi-modal Large Language Models (MLLMs) has advanced workflow automation; however, existing research mainly targets performance upper bounds in static environments, overlooking robustness for stochastic real-world deployment. We identify three key challenges: dynamic task scheduling, active exploration under uncertainty, and continuous learning from experience. To bridge this gap, we introduce a dynamic evaluation environment that simulates a "trainee" agent continuously exploring a novel setting. Unlike traditional benchmarks, it evaluates agents along three dimensions: (1) context-aware scheduling for streaming tasks with varying priorities; (2) prudent information acquisition to reduce hallucination via active exploration; and (3) continuous evolution by distilling generalized strategies from rule-based, dynamically generated tasks. Experiments show that cutting-edge agents have significant deficiencies in dynamic environments, especially in active exploration and continual learning. Our work establishes a framework for assessing agent reliability, shifting evaluation from static tests to realistic, production-oriented scenarios.

   @article{fu2026agentfirstday,
     title = {The Agent's First Day: Benchmarking Learning, Exploration, and Scheduling in the Workplace Scenarios},
     author = {Fu*, Daocheng and Mei*, Jianbiao and Wu*, Rong and Yang, Xuemeng and Xu, Jia and Wang, Ding and Cai, Pinlong and Liu, Yong and Wen, Licheng and Shi, Botian},
     journal = {Findings of the Association for Computational Linguistics (ACL)},
     year = {2026},
   }

2. ACL Findings

   Towards Self-Evolving Agents: Enabling Autonomy through Interactive Experience Refinement

   Cheng Yang*, Xuemeng Yang*, Licheng Wen*, and 9 more authors

   Findings of the Association for Computational Linguistics (ACL), 2026

   Abs arXiv Bib Code

   Large Language Models have demonstrated remarkable capabilities across diverse domains, yet significant challenges persist when deploying them as AI agents for real-world long-horizon tasks. Existing LLM agents suffer from a critical limitation: they are test-time static and cannot learn from experience, lacking the ability to accumulate knowledge and continuously improve on the job. To address this challenge, we propose MUSE, a novel agent framework that introduces an experience-driven, self-evolving system centered around a hierarchical Memory Module. MUSE organizes diverse levels of experience and leverages them to plan and execute long-horizon tasks across multiple applications. After each sub-task execution, the agent autonomously reflects on its trajectory, converting the raw trajectory into structured experience and integrating it back into the Memory Module. This mechanism enables the agent to evolve beyond its static pretrained parameters, fostering continuous learning and self-evolution. We evaluate MUSE on the long-horizon productivity benchmark TAC. It achieves new SOTA performance by a significant margin using only a lightweight Gemini-2.5 Flash model. Sufficient Experiments demonstrate that as the agent autonomously accumulates experience, it exhibits increasingly superior task completion capabilities, as well as robust continuous learning and self-evolution capabilities. Moreover, the accumulated experience from MUSE exhibits strong generalization properties, enabling zero-shot improvement on new tasks.

   @article{yang2026muse,
     title = {Towards Self-Evolving Agents: Enabling Autonomy through Interactive Experience Refinement},
     author = {Yang*, Cheng and Yang*, Xuemeng and Wen*, Licheng and Fu, Daocheng and Mei, Jianbiao and Wu, Rong and Cai, Pinlong and Shen, Yufan and Deng, Nianchen and Shi, Botian and Qiao, Yu and Li, Haifeng},
     journal = {Findings of the Association for Computational Linguistics (ACL)},
     year = {2026},
   }

2025

1. Preprint

   O^2-Searcher: A Searching-based Agent Model for Open-Domain Open-Ended Question Answering

   Jianbiao Mei, Tao Hu, Daocheng Fu, and 11 more authors

   arXiv preprint arXiv:2505.16582, 2025

   Abs arXiv Bib Code

   Large Language Models (LLMs), despite their advancements, are fundamentally limited by their static parametric knowledge, hindering performance on tasks requiring open-domain up-to-date information. While enabling LLMs to interact with external knowledge environments is a promising solution, current efforts primarily address closed-end problems. Open-ended questions, which characterized by lacking a standard answer or providing non-unique and diverse answers, remain underexplored. To bridge this gap, we present O2-Searcher, a novel search agent leveraging reinforcement learning to effectively tackle both open-ended and closed-ended questions in the open domain. O2-Searcher leverages an efficient, locally simulated search environment for dynamic knowledge acquisition, effectively decoupling the external world knowledge from model’s sophisticated reasoning processes. It employs a unified training mechanism with meticulously designed reward functions, enabling the agent to identify problem types and adapt different answer generation strategies. Furthermore, to evaluate performance on complex open-ended tasks, we construct O-QA, a high-quality benchmark featuring 300 manually curated, multi-domain open-ended questions with associated web page caches. Extensive experiments show that O2-Searcher, using only a 3B model, significantly surpasses leading LLM agents on O-QA. It also achieves SOTA results on various closed-ended QA benchmarks against similarly-sized models, while performing on par with much larger ones.

   @article{mei2025o2searchersearchingbasedagentmodel,
     title = {O$^2$-Searcher: A Searching-based Agent Model for Open-Domain Open-Ended Question Answering},
     author = {Mei, Jianbiao and Hu, Tao and Fu, Daocheng and Wen, Licheng and Yang, Xuemeng and Wu, Rong and Cai, Pinlong and Cai, Xinyu and Gao, Xing and Yang, Yu and Xie, Chengjun and Shi, Botian and Liu, Yong and Qiao, Yu},
     journal = {arXiv preprint arXiv:2505.16582},
     year = {2025},
   }

2. 

   DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving

   Xuemeng Yang*, Licheng Wen*, Yukai Ma*, and 11 more authors

   2025 IEEE/CVF International Conference on Computer Vision (ICCV), 2025

   Abs arXiv Bib Demo Code

   This paper presetns DriveArena, the first high-fidelity closed-loop simulation system designed for driving agents navigating in real scenarios. DriveArena features a flexible, modular architecture, allowing for the seamless interchange of its core components: Traffic Manager, a traffic simulator capable of generating realistic traf- fic flow on any worldwide street map, and World Dreamer, a high-fidelity conditional generative model with infinite autoregression. This powerful synergy empowers any driving agent capable of processing real-world images to navigate in DriveArena simulated environment. The agent perceives its surroundings through images generated by World Dreamer and output trajectories; then these trajectories are fed into Traffic Manager, achieving realistic interactions with other vehicles and producing a new scene lay- out. Finally, the latest scene layout is relayed back into World Dreamer, perpetuating the simulation cycle. This iterative process fosters closed-loop exploration within a highly realistic environment, providing a valuable platform for developing and evaluating driving agents across diverse and challenging scenarios. DriveArena signifies a substantial leap forward in leveraging generative image data for the driving simulatior, opening insights for closed-loop autonomous driving.

   @article{yang2024drivearena,
     title = {DriveArena: A Closed-loop Generative Simulation Platform for Autonomous Driving},
     author = {Yang*, Xuemeng and Wen*, Licheng and Ma*, Yukai and Mei*, Jianbiao and Li*, Xin and Wei*, Tiantian and Lei, Wenjie and Fu, Daocheng and Cai, Pinlong and Dou, Min and Shi, Botian and He, Liang and Liu, Yong and Qiao, Yu},
     journal = {2025 IEEE/CVF International Conference on Computer Vision (ICCV)},
     year = {2025},
     demo = {https://pjlab-adg.github.io/DriveArena/},
   }

2024

1. ICLR

   DiLu: A Knowledge-Driven Approach to Autonomous Driving with Large Language Models

   Licheng Wen*, Daocheng Fu*, Xin Li*, and 7 more authors

   In The Eleventh International Conference on Learning Representations (ICLR), 2024

   Abs arXiv Bib Demo Code

   Recent advancements in autonomous driving have relied on data-driven approaches, which are widely adopted but face challenges including dataset bias, overfitting, and uninterpretability. Drawing inspiration from the knowledge-driven nature of human driving, we explore the question of how to instill similar capabilities into autonomous driving systems and summarize a paradigm that integrates an interactive environment, a driver agent, as well as a memory component to address this question. Leveraging large language models with emergent abilities, we propose the DiLu framework, which combines a Reasoning and a Reflection module to enable the system to perform decision-making based on common-sense knowledge and evolve continuously. Extensive experiments prove DiLu’s capability to accumulate experience and demonstrate a significant advantage in generalization ability over reinforcement learning-based methods. Moreover, DiLu is able to directly acquire experiences from real-world datasets which highlights its potential to be deployed on practical autonomous driving systems. To the best of our knowledge, we are the first to instill knowledge-driven capability into autonomous driving systems from the perspective of how humans drive.

   @inproceedings{wen2023dilu,
     title = {DiLu: A Knowledge-Driven Approach to Autonomous Driving with Large Language Models},
     author = {Wen*, Licheng and Fu*, Daocheng and Li*, Xin and Cai, Xinyu and Ma, Tao and Cai, Pinlong and Dou, Min and Shi, Botian and He, Liang and Qiao, Yu},
     booktitle = {The Eleventh International Conference on Learning Representations (ICLR)},
     year = {2024},
     demo = {https://pjlab-adg.github.io/DiLu/},
   }

2. 

   On the Road with GPT-4V(ision): Early Explorations of Visual-Language Model on Autonomous Driving

   Licheng Wen*, Xuemeng Yang*, Daocheng Fu*, and 14 more authors

   In ICLR 2024 Workshop on Large Language Model (LLM) Agents, 2024

   Abs arXiv Bib Code

   The development of autonomous driving technology depends on merging perception, decision, and control systems. Traditional strategies have struggled to understand complex driving environments and other road users’ intent. This bottleneck, especially in constructing common sense reasoning and nuanced scene understanding, affects the safe and reliable operations of autonomous vehicles. The introduction of Visual Language Models (VLM) opens up possibilities for fully autonomous driving. This report evaluates the potential of GPT-4V(ision), the latest state-of-the-art VLM, as an autonomous driving agent. The evaluation primarily assesses the model’s ultimate ability to act as a driving agent under varying conditions, while also considering its capacity to understand driving scenes and make decisions. Findings show that GPT-4V outperforms existing systems in scene understanding and causal reasoning. It has the potential in handling unexpected scenarios, understanding intentions, and making informed decisions. However, limitations remain in direction determination, traffic light recognition, vision grounding, and spatial reasoning tasks, highlighting the need for further research.

   @inproceedings{wen2024road,
     title = {On the Road with GPT-4V(ision): Early Explorations of Visual-Language Model on Autonomous Driving},
     author = {Wen*, Licheng and Yang*, Xuemeng and Fu*, Daocheng and Wang*, Xiaofeng and Cai, Pinlong and Li, Xin and Ma, Tao and Li, Yingxuan and Xu, Linran and Shang, Dengke and Zhu, Zheng and Sun, Shaoyan and Bai, Yeqi and Cai, Xinyu and Dou, Min and Hu, Shuanglu and Shi, Botian},
     booktitle = {ICLR 2024 Workshop on Large Language Model (LLM) Agents},
     year = {2024},
   }

2023

1. ITSC

   LimSim: A Long-term Interactive Multi-scenario Traffic Simulator

   Licheng Wen*, Daocheng Fu*, Song Mao, and 3 more authors

   In 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), 2023

   Abs arXiv Bib Video Code

   With the growing popularity of digital twin and autonomous driving in transportation, the demand for simulation systems capable of generating high-fidelity and reliable scenarios is increasing. Existing simulation systems suffer from a lack of support for different types of scenarios, and the vehicle models used in these systems are too simplistic. Thus, such systems fail to represent driving styles and multi-vehicle interactions, and struggle to handle corner cases in the dataset. In this paper, we propose LimSim, the Long-term Interactive Multi-scenario traffic Simulator, which aims to provide a long-term continuous simulation capability under the urban road network. LimSim can simulate fine-grained dynamic scenarios and focus on the diverse interactions between multiple vehicles in the traffic flow. This paper provides a detailed introduction to the framework and features of the LimSim, and demonstrates its performance through case studies and experiments. LimSim is now open source on GitHub.

   @inproceedings{wen2023limsim,
     title = {LimSim: A Long-term Interactive Multi-scenario Traffic Simulator},
     author = {Wen*, Licheng and Fu*, Daocheng and Mao, Song and Cai, Pinlong and Dou, Min and Li, Yikang},
     year = {2023},
     booktitle = {2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC)},
     video = {https://www.youtube.com/playlist?list=PLNeNtm096CAyYD1JJnkQ4gMaoFSdFLn2y},
   }

2022

1. 

   CL-MAPF: Multi-Agent Path Finding for Car-Like robots with kinematic and spatiotemporal constraints

   Licheng Wen, Yong Liu, and Hongliang Li

   Robotics and Autonomous Systems, 2022

   Abs arXiv Bib Video Code

   Multi-Agent Path Finding has been widely studied in the past few years due to its broad application in the field of robotics and AI. However, previous solvers rely on several simplifying assumptions. This limits their applicability in numerous real-world domains that adopt nonholonomic car-like agents rather than holonomic ones. In this paper, we give a mathematical formalization of the Multi-Agent Path Finding for Car-Like robots (CL-MAPF) problem. We propose a novel hierarchical search-based solver called Car-Like Conflict-Based Search to address this problem. It applies a body conflict tree to address collisions considering the shapes of the agents. We introduce a new algorithm called Spatiotemporal Hybrid-State A* as the single-agent planner to generate agents’ paths satisfying both kinematic and spatiotemporal constraints. We also present a sequential planning version of our method, sacrificing a small amount of solution quality to achieve a significant reduction in runtime. We compare our method with two baseline algorithms on a dedicated benchmark and validate it in real-world scenarios. The experiment results show that the planning success rate of both baseline algorithms is below 50% for all six scenarios, while our algorithm maintains that of over 98%. It also gives clear evidence that our algorithm scales well to 100 agents in 300 m × 300 m scenario and is able to produce solutions that can be directly applied to Ackermann-steering robots in the real world. The benchmark and source code are released in https://github.com/APRIL-ZJU/CL-CBS. The video of the experiments can be found on YouTube.

   @article{10.1016/j.robot.2021.103997,
     title = {CL-MAPF: Multi-Agent Path Finding for Car-Like robots with kinematic and spatiotemporal constraints},
     journal = {Robotics and Autonomous Systems},
     volume = {150},
     pages = {103997},
     year = {2022},
     issn = {0921-8890},
     doi = {https://doi.org/10.1016/j.robot.2021.103997},
     url = {https://www.sciencedirect.com/science/article/pii/S0921889021002530},
     author = {Wen, Licheng and Liu, Yong and Li, Hongliang},
     keywords = {Multi-agent systems, Path planning, Mobile robots},
     video = {https://www.youtube.com/watch?v=KThsX04ABvc},
   }