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},
     homepage = {true}
   }

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},
     homepage = {true}
   }

3. 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

   Transactions on Machine Learning Research (TMLR), 2026

   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 = {Transactions on Machine Learning Research (TMLR)},
     year = {2026},
     homepage = {true},
   }

2025

1. 

   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/},
     homepage = {true}
   }