BP
Baolin Peng
cs.CLcs.AIcs.LGcs.CVcs.NEcs.HCcs.IRcs.ROcs.SEAlgorithms
On Valency
published · living versionsW_swcb792r·v1 · currentpublished
Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
with Zhaoyang Wang, Qianhui Wu, Haorui Wang, Simon Yu +7
1 version
Preprints & journals
91 papers in the corpus · 2015–2026Agent World Model in the Wild: Computer-Use Agent Environment Synthesis2610.26828v1 · Zhaoyang Wang, Qianhui Wu, Haorui Wang et al.2026 · 0 citationsarXivon Valency
Self-Evolving Coding Agents via Harness-Policy Optimization2610.22791v1 · Zhepei Wei, Wenlin Yao, Baolin Peng et al.2026 · 0 citationsarXiv
AutoEnvScaling: Automating the Data Flywheel with Terminal Agents2610.14163v1 · Simon Yu, Baolin Peng, Bo Liu et al.2026 · 0 citationsarXiv
SecureVibe: Making Vibe Coding More Secure2609.38606v1 · Danqing Wang, Baolin Peng, Zhepei Wei et al.2026 · 0 citationsarXiv
Certified Long-Horizon Code Agent Evolution via Validation-Gated Skill Optimization2609.32990v1 · Yifan Wang, Hao Cheng, Xiaomin Li et al.2026 · 0 citationsarXiv
ExecCritic: Learn to Test, Test to Improve for Coding Agents2609.09133v1 · Leitian Tao, Baolin Peng, Haorui Wang et al.2026 · 0 citationsarXiv
FrogNano: Training a 4B Coding Agent via Online Task Synthesis2609.07925v5 · Minseon Kim, Zhengyan Shi, Emiliano Penaloza et al.2026 · 0 citationsarXiv
OpenForgeRL: Train Harness-native Agents in Any Environment2607.21557v3 · Xiao Yu, Baolin Peng, Ruize Xu et al.2026 · 0 citationsarXiv
OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents2606.02031v2 · Rui Yang, Qianhui Wu, Yuxi Chen et al.2026 · 0 citationsarXiv
Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior2605.26797v2 · Zeyi Huang, Xuehai He, LiLiang Ren et al.2026 · 0 citationsarXiv
Orchard: An Open-Source Agentic Modeling Framework2605.15040v3 · Baolin Peng, Wenlin Yao, Qianhui Wu et al.2026 · 0 citationsarXiv
Synthetic Computers at Scale for Long-Horizon Productivity Simulation2604.28181v1 · Tao Ge, Baolin Peng, Hao Cheng et al.2026 · 0 citationsarXiv
WebXSkill: Skill Learning for Autonomous Web Agents2604.13318v2 · Zhaoyang Wang, Qianhui Wu, Xuchao Zhang et al.2026 · 0 citationsarXiv
The Tool Illusion: Rethinking Tool Use in Web Agents2604.03465v2 · Renze Lou, Baolin Peng, Wenlin Yao et al.2026 · 0 citationsarXiv
GUI-Libra: Training Native GUI Agents to Reason and Act with Action-aware Supervision and Partially Verifiable RL2602.22190v2 · Rui Yang, Qianhui Wu, Zhaoyang Wang et al.2026 · 0 citationsarXiv
Multimodal Reinforcement Learning with Adaptive Verifier for AI Agents2512.03438v3 · Reuben Tan, Baolin Peng, Zhengyuan Yang et al.2025 · 0 citationsarXiv
ThetaEvolve: Test-time Learning on Open Problems2511.23473v1 · Yiping Wang, Shao-Rong Su, Zhiyuan Zeng et al.2025 · 0 citationsarXiv
SynthAgent: Adapting Web Agents with Synthetic Supervision2511.06101v3 · Zhaoyang Wang, Yiming Liang, Xuchao Zhang et al.2025 · 0 citationsarXiv
Dyna-Mind: Learning to Simulate from Experience for Better AI Agents2510.09577v1 · Xiao Yu, Baolin Peng, Michel Galley et al.2025 · 0 citationsarXiv
SimulatorArena: Are User Simulators Reliable Proxies for Multi-Turn Evaluation of AI Assistants?2510.05444v2 · Yao Dou, Michel Galley, Baolin Peng et al.2025 · 1 citationarXiv
Decoder-Hybrid-Decoder Architecture for Efficient Reasoning with Long Generation2507.06607v3 · Liliang Ren, Congcong Chen, Haoran Xu et al.2025 · 0 citationsarXiv
GUI-Actor: Coordinate-Free Visual Grounding for GUI Agents2506.03143v1 · Qianhui Wu, Kanzhi Cheng, Rui Yang et al.2025 · 4 citationsarXiv
Dyna-Think: Synergizing Reasoning, Acting, and World Model Simulation in AI Agents2506.00320v3 · Xiao Yu, Baolin Peng, Ruize Xu et al.2025 · 0 citationsarXiv
Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math2504.21233v1 · Haoran Xu, Baolin Peng, Hany Awadalla et al.2025 · 0 citationsarXiv
Reinforcement Learning for Reasoning in Large Language Models with One Training Example2504.20571v3 · Yiping Wang, Qing Yang, Zhiyuan Zeng et al.2025 · 5 citationsarXiv
Magma: A Foundation Model for Multimodal AI Agents2502.13130v1 · Jianwei Yang, Reuben Tan, Qianhui Wu et al.2025 · 33 citationsarXiv
On the Emergence of Thinking in LLMs I: Searching for the Right Intuition2502.06773v1 · Guanghao Ye, Khiem Duc Pham, Xinzhi Zhang et al.2025 · 0 citationsarXiv
CollabLLM: From Passive Responders to Active Collaborators2502.00640v3 · Shirley Wu, Michel Galley, Baolin Peng et al.2025 · 0 citationsarXiv
Latent Action Pretraining from Videos2410.11758v2 · Seonghyeon Ye, Joel Jang, Byeongguk Jeon et al.2024 · 1 citationarXiv
Towards Self-Improvement of LLMs via MCTS: Leveraging Stepwise Knowledge with Curriculum Preference Learning2410.06508v1 · Xiyao Wang, Linfeng Song, Ye Tian et al.2024 · 0 citationsarXiv
ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning2410.02052v5 · Xiao Yu, Baolin Peng, Vineeth Vajipey et al.2024 · 0 citationsarXiv
SIaM: Self-Improving Code-Assisted Mathematical Reasoning of Large Language Models2408.15565v1 · Dian Yu, Baolin Peng, Ye Tian et al.2024 · 2 citationsarXiv
Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning2407.00617v4 · Yuheng Zhang, Dian Yu, Baolin Peng et al.2024 · 1 citationarXiv
Self-Tuning: Instructing LLMs to Effectively Acquire New Knowledge through Self-Teaching2406.06326v5 · Xiaoying Zhang, Baolin Peng, Ye Tian et al.2024 · 0 citationsarXiv
Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing2404.12253v2 · Ye Tian, Baolin Peng, Linfeng Song et al.2024 · 7 citationsarXiv
Entropy Guided Extrapolative Decoding to Improve Factuality in Large Language Models2404.09338v1 · Souvik Das, Lifeng Jin, Linfeng Song et al.2024 · 2 citationsarXiv
Conceptual and Unbiased Reasoning in Language Models2404.00205v1 · Ben Zhou, Hongming Zhang, Sihao Chen et al.2024 · 0 citationsarXiv
Self-Consistency Boosts Calibration for Math Reasoning2403.09849v1 · Ante Wang, Linfeng Song, Ye Tian et al.2024 · 3 citationsarXiv
Collaborative decoding of critical tokens for boosting factuality of large language models2402.17982v1 · Lifeng Jin, Baolin Peng, Linfeng Song et al.2024 · 1 citationarXiv
Fine-Grained Self-Endorsement Improves Factuality and Reasoning2402.15631v1 · Ante Wang, Linfeng Song, Baolin Peng et al.2024 · 1 citationarXiv
Self-Alignment for Factuality: Mitigating Hallucinations in LLMs via Self-Evaluation2402.09267v2 · Xiaoying Zhang, Baolin Peng, Ye Tian et al.2024 · 23 citationsACL2024 Main
Sub-Sentence Encoder: Contrastive Learning of Propositional Semantic Representations2311.04335v1 · Sihao Chen, Hongming Zhang, Tong Chen et al.2023 · 4 citationsarXiv
Teaching Language Models to Self-Improve through Interactive Demonstrations2310.13522v2 · Xiao Yu, Baolin Peng, Michel Galley et al.2023 · 4 citationsarXiv
Stabilizing RLHF through Advantage Model and Selective Rehearsal2309.10202v1 · Baolin Peng, Linfeng Song, Ye Tian et al.2023 · 2 citationsarXiv
Evaluating the Instruction-Following Robustness of Large Language Models to Prompt Injection2308.10819v3 · Zekun Li, Baolin Peng, Pengcheng He et al.2023 · 28 citationsarXiv
Self-Checker: Plug-and-Play Modules for Fact-Checking with Large Language Models2305.14623v2 · Miaoran Li, Baolin Peng, Michel Galley et al.2023 · 35 citationsarXiv
SGP-TOD: Building Task Bots Effortlessly via Schema-Guided LLM Prompting2305.09067v1 · Xiaoying Zhang, Baolin Peng, Kun Li et al.2023 · 21 citationsarXiv
ArK: Augmented Reality with Knowledge Interactive Emergent Ability2305.00970v1 · Qiuyuan Huang, Jae Sung Park, Abhinav Gupta et al.2023 · 2 citationsarXiv
Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models2304.09842v3 · Pan Lu, Baolin Peng, Hao Cheng et al.2023 · 142 citationsarXiv
Instruction Tuning with GPT-42304.03277v1 · Baolin Peng, Chunyuan Li, Pengcheng He et al.2023 · 175 citationsarXiv
Career total: 134 works. 91 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.