Saved in:
Bibliographic Details
Main Authors: Kalyan, Vivek, Andrews, Martin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.24126
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915582398431232
author Kalyan, Vivek
Andrews, Martin
author_facet Kalyan, Vivek
Andrews, Martin
contents Large Language Model (LLM) agents can leverage multiple turns and tools to solve complex tasks, with prompt-based approaches achieving strong performance. This work demonstrates that Reinforcement Learning (RL) can push capabilities significantly further by learning from experience. Through experiments on a legal document search benchmark, we show that our RL-trained 14 Billion parameter model outperforms frontier class models (85% vs 78% accuracy). In addition, we explore turn-restricted regimes, during training and at test-time, that show these agents achieve better results if allowed to operate over longer multi-turn horizons.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24126
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Reinforcement Learning for Long-Horizon Multi-Turn Search Agents
Kalyan, Vivek
Andrews, Martin
Computation and Language
Large Language Model (LLM) agents can leverage multiple turns and tools to solve complex tasks, with prompt-based approaches achieving strong performance. This work demonstrates that Reinforcement Learning (RL) can push capabilities significantly further by learning from experience. Through experiments on a legal document search benchmark, we show that our RL-trained 14 Billion parameter model outperforms frontier class models (85% vs 78% accuracy). In addition, we explore turn-restricted regimes, during training and at test-time, that show these agents achieve better results if allowed to operate over longer multi-turn horizons.
title Reinforcement Learning for Long-Horizon Multi-Turn Search Agents
topic Computation and Language
url https://arxiv.org/abs/2510.24126