paper-with-me

홈 › Papers

LAM SIMULATOR: Advancing Data Generation for Large Action Model Training via Online Exploration and Trajectory Feedback

2025-06-02 · Thai Hoang, Kung-Hsiang Huang, Shirley Kokane, JianGuo Zhang, Zuxin Liu, Ming Zhu, Jake Grigsby, Tian Lan, Michael S Ryoo, Chien-Sheng Wu, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong, Juan Carlos Niebles

Large Action Models (LAMs) for AI Agents offer incredible potential but face challenges due to the need for high-quality training data, especially for multi-steps tasks that involve planning, executing tool calls, and responding to feedback. To address these issues, we present LAM SIMULATOR, a comprehensive framework designed for online exploration of agentic tasks with high-quality feedback. Our framework features a dynamic task query generator, an extensive collection of tools, and an interactive environment where Large Language Model (LLM) Agents can call tools and receive real-time feedback. This setup enables LLM Agents to explore and solve tasks autonomously, facilitating the discovery of multiple approaches to tackle any given task. The resulting action trajectory data are then used to create high-quality training datasets for LAMs. Our experiments on popular agentic benchmarks, ToolBench and CRMArena, highlight the effectiveness of LAM SIMULATOR: models trained with self-generated datasets using our framework achieve significant performance gains, up to a 49.3\% improvement over their original baselines. LAM SIMULATOR requires minimal human input during dataset creation, highlighting LAM SIMULATOR's efficiency and effectiveness in speeding up development of AI agents.

📄 PDF Abstract BibTeX arXiv:2506.02298

Code (0)

등록된 구현이 없습니다.

Tasks

Large Language Model

Similar Papers 제목 키워드 기반

Kinema4D: Kinematic 4D World Modeling for Spatiotemporal Embodied Simulation

2026-03-17 · Mutian Xu, Tianbao Zhang, Tianqi Liu, Zhaoxi Chen 외 arxiv

Simulating robot-world interactions is a cornerstone of Embodied AI. Recently, a few works have shown promise in leveraging video generations to transcend the rigid visual/physical constraints of traditional simulators. …

NASimJax: A GPU-Accelerated Policy Learning Framework for Penetration Testing

2026-03-20 · Raphael Simon, José Carrasquel, Wim Mees, Pieter Libin arxiv

Penetration testing, the practice of simulating cyberattacks to identify vulnerabilities, is a complex sequential decision-making task that is inherently partially observable and features large action spaces. Training re…

Reinforcement Learning

Grounded Gesture Generation: Language, Motion, and Space

2025-07-06 · Anna Deichler, Jim O'Regan, Teo Guichoux, David Johansson 외 arxiv

Human motion generation has advanced rapidly in recent years, yet the critical problem of creating spatially grounded, context-aware gestures has been largely overlooked. Existing models typically specialize either in de…

Synthetic Data GenerationGesture Generation

FlightForge: Advancing UAV Research with Procedural Generation of High-Fidelity Simulation and Integrated Autonomy

2025-02-07 · David Čapek, Jan Hrnčíř, Tomáš Báča, Jakub Jirkal 외

Robotic simulators play a crucial role in the development and testing of autonomous systems, particularly in the realm of Uncrewed Aerial Vehicles (UAV). However, existing simulators often lack high-level autonomy, hinde…

Autonomous NavigationCollision Avoidance

LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks

2025-05-31 · Yi Yang, Jiaxuan Sun, Siqi Kou, Yihan Wang 외

Real-world embodied agents face long-horizon tasks, characterized by high-level goals demanding multi-step solutions beyond single actions. Successfully navigating these requires both high-level task planning (i.e., deco…

Task PlanningVision-Language-Action