AI

Reinforcement Learning for Long-Horizon Interactive LLM Agents

1 Mins read

Interactive digital agents (IDAs) leverage APIs of stateful digital environments to perform tasks in response to user requests. While IDAs powered by instruction-tuned large language models (LLMs) can react to feedback from interface invocations in multi-step exchanges, they have not been trained in their respective digital environments. Prior methods accomplish less than half of tasks in sophisticated benchmarks such as AppWorld. We present a reinforcement learning (RL) approach that trains IDAs directly in their target environments. We formalize this training as a partially observable Markov decision process and derive LOOP, a data- and memory-efficient variant of proximal policy optimization. LOOP uses no value network and maintains exactly one copy of the underlying LLM in memory, making its implementation straightforward and as memory-efficient as fine-tuning a single LLM. A 32-billion-parameter agent trained with LOOP in the AppWorld environment outperforms the much larger OpenAI o1 agent by 9 percentage points (15% relative). To our knowledge, this is the first reported application of RL to IDAs that interact with a stateful, multi-domain, multi-app environment via direct API calls. Our analysis sheds light on the effectiveness of RL in this area, showing that the agent learns to consult the API documentation, avoid unwarranted assumptions, minimize confabulation, and recover from setbacks.


Source link

Related posts
AI

Vintix: Scaling In-Context Reinforcement Learning for Generalist AI Agents

2 Mins read
Developing AI systems that learn from their surroundings during execution involves creating models that adapt dynamically based on new information. In-Context Reinforcement…
AI

LLMDet: How Large Language Models Enhance Open-Vocabulary Object Detection

3 Mins read
Open-vocabulary object detection (OVD) aims to detect arbitrary objects with user-provided text labels. Although recent progress has enhanced zero-shot detection ability, current…
AI

Zyphra Introduces the Beta Release of Zonos: A Highly Expressive TTS Model with High Fidelity Voice Cloning

3 Mins read
Text-to-speech (TTS) technology has made significant strides in recent years, but challenges remain in creating natural, expressive, and high-fidelity speech synthesis. Many…

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *