9 papers
LLM Value Functions for Credit Assignment in Agentic RL
Kaijie Zhu, Weixiang Yan, Omar Attia +3
Reinforcement learning (RL) for large language model (LLM) agents remains challenging in long-horizon tasks with sparse outcome rewards, where assigning credit to individual action…
RLTLDR: Self-improvement by Internalizing Self-generated Feedback
Michael Kirchhof, Eleonora Gualdoni, Andrew Szot +6
The common paradigm of reinforcement learning with verifiable rewards (RLVR) is to let agents make multiple attempts at a task, and optimize towards the successful ones. This becom…
Expanding LLM Agent Boundaries with Strategy-Guided Exploration
Andrew Szot, Michael Kirchhof, Omar Attia +1
Reinforcement learning (RL) has demonstrated notable success in post-training large language models (LLMs) as agents for tasks such as computer use, tool calling, and coding. Howev…
ASTRA-bench: Evaluating Tool-Use Agent Reasoning and Action Planning with Personal User Context
Zidi Xiu, David Q. Sun, Kevin Cheng +9
Next-generation AI must manage vast personal data, diverse tools, and multi-step reasoning, yet most benchmarks remain context-free and single-turn. We present ASTRA-bench (Assista…
UltraCUA: A Foundation Model for Computer Use Agents with Hybrid Action
Yuhao Yang, Zhen Yang, Zi-Yi Dou +10
Computer-use agents face a fundamental limitation. They rely exclusively on primitive GUI actions (click, type, scroll), creating brittle execution chains prone to cascading failur…
GRACE: A Language Model Framework for Explainable Inverse Reinforcement Learning
Silvia Sapora, Devon Hjelm, Alexander Toshev +2
Inverse Reinforcement Learning aims to recover reward models from expert demonstrations, but traditional methods yield black-box models that are difficult to interpret and debug. I…
Ferret-UI Lite: Lessons from Building Small On-Device GUI Agents
Zhen Yang, Zi-Yi Dou, Di Feng +13
Developing autonomous agents that effectively interact with Graphic User Interfaces (GUIs) remains a challenging open problem, especially for small on-device models. In this paper,…
Scaling Synthetic Task Generation for Agents via Exploration
Ram Ramrakhya, Andrew Szot, Omar Attia +6
Post-Training Multimodal Large Language Models (MLLMs) to build interactive agents holds promise across domains such as computer-use, web navigation, and robotics. A key challenge…
From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessons
Andrew Szot, Bogdan Mazoure, Omar Attia +6
We examine the capability of Multimodal Large Language Models (MLLMs) to tackle diverse domains that extend beyond the traditional language and vision tasks these models are typica…