A Subgoal-driven Framework for Improving Long-Horizon LLM Agents

• URL: http://arxiv.org/abs/2603.19685v1
• Date: Fri, 20 Mar 2026 06:32:16 GMT
• Title: A Subgoal-driven Framework for Improving Long-Horizon LLM Agents
• Abstract summary: Large language model (LLM)-based agents have emerged as powerful autonomous controllers for digital environments.<n>We present MiRA, an RL training framework that uses dense, milestone-based reward signals.<n>Our findings demonstrate that combining explicit inference-time planning with milestone-based rewards significantly improves an agent's long-horizon capabilities.
• Score: 12.411629679174183
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model (LLM)-based agents have emerged as powerful autonomous controllers for digital environments, including mobile interfaces, operating systems, and web browsers. Web navigation, for example, requires handling dynamic content and long sequences of actions, making it particularly challenging. Existing LLM-based agents struggle with long-horizon planning in two main ways. During online execution, they often lose track as new information arrives, lacking a clear and adaptive path toward the final goal. This issue is further exacerbated during reinforcement learning (RL) fine-tuning, where sparse and delayed rewards make it difficult for agents to identify which actions lead to success, preventing them from maintaining coherent reasoning over extended tasks. To address these challenges, we propose two contributions. First, we introduce an agent framework that leverages proprietary models for online planning through subgoal decomposition. Second, we present MiRA (Milestoning your Reinforcement Learning Enhanced Agent), an RL training framework that uses dense, milestone-based reward signals. The real-time planning mechanism improves proprietary models such as Gemini by approximately a 10% absolute increase in success rate (SR) on the WebArena-Lite benchmark. Meanwhile, applying MiRA to the open Gemma3-12B model increases its success rate from 6.4% to 43.0%. This performance surpasses proprietary systems such as GPT-4-Turbo (17.6%) and GPT-4o (13.9%), as well as the previous open-model state of the art, WebRL (38.4%). Overall, our findings demonstrate that combining explicit inference-time planning with milestone-based rewards significantly improves an agent's long-horizon capabilities, paving the way for more robust and general-purpose autonomous systems.

Related papers

• M$^2$: Dual-Memory Augmentation for Long-Horizon Web Agents via Trajectory Summarization and Insight Retrieval [64.06936170117943]
  M$2$ is a training-free, memory-augmented framework designed to optimize context efficiency and decision-making.<n>Our approach incorporates a dual-tier memory mechanism that synergizes Dynamic Trajectory Summarization (Internal Memory) to compress verbose interaction history into concise state updates, and Insight Retrieval Augmentation (External Memory) to guide the agent with actionable guidelines retrieved from an offline insight bank.
  arXiv  Detail & Related papers  (2026-02-28T06:59:51Z)
• OpAgent: Operator Agent for Web Navigation [23.928869500029432]
  We develop an online interaction environment and fine-tune the Vision-Language Model (VLM) using a specialized RL pipeline.<n>We introduce a Hybrid Reward Mechanism that combines a ground-truth-agnostic WebJudge for holistic outcome assessment and a Rule-based Decision Tree (RDT) for progress reward.<n> Notably, our RL-enhanced model achieves a 38.1% success rate (pass@5) on WebArena, outperforming all existing monolithic baselines.
  arXiv  Detail & Related papers  (2026-02-14T02:33:55Z)
• Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agents [49.119608399413806]
  Large language models (LLMs) are increasingly deployed as autonomous agents for multi-turn decision-making tasks.<n>This paper introduces Cog, a framework that trains agents to dynamically adapt cognitive depth at each step.<n> Experiments on ALFWorld and ScienceWorld demonstrate that Cog achieves state-of-the-art performance with superior efficiency.
  arXiv  Detail & Related papers  (2026-02-13T06:52:09Z)
• WebAnchor: Anchoring Agent Planning to Stabilize Long-Horizon Web Reasoning [82.12501258760814]
  Large Language Model(LLM)-based agents have shown strong capabilities in web information seeking.<n>Plan anchor is where the first reasoning step disproportionately impacts downstream behavior in long-horizon web reasoning tasks.<n>We propose Anchor-GRPO, a two-stage RL framework that decouples planning and execution.
  arXiv  Detail & Related papers  (2026-01-06T16:36:40Z)
• Stronger Together: On-Policy Reinforcement Learning for Collaborative LLMs [20.084201133669534]
  Multi-agent systems (MAS) and reinforcement learning (RL) are widely used to enhance the agentic capabilities of large language models (LLMs)<n>Applying on-policy RL to MAS remains underexplored and presents unique challenges.<n>We propose AT-GRPO, which includes (i) an agent- and turn-wise grouped RL algorithm tailored to MAS and (ii) a training system that supports both single- and multi-policy regimes.
  arXiv  Detail & Related papers  (2025-10-13T06:55:09Z)
• Agentic Reinforcement Learning with Implicit Step Rewards [92.26560379363492]
  Large language models (LLMs) are increasingly developed as autonomous agents using reinforcement learning (agentic RL)<n>We introduce implicit step rewards for agentic RL (iStar), a general credit-assignment strategy that integrates seamlessly with standard RL algorithms.<n>We evaluate our method on three challenging agent benchmarks, including WebShop and VisualSokoban, as well as open-ended social interactions with unverifiable rewards in SOTOPIA.
  arXiv  Detail & Related papers  (2025-09-23T16:15:42Z)
• xLAM: A Family of Large Action Models to Empower AI Agent Systems [111.5719694445345]
  We release xLAM, a series of large action models designed for AI agent tasks. xLAM consistently delivers exceptional performance across multiple agent ability benchmarks.
  arXiv  Detail & Related papers  (2024-09-05T03:22:22Z)
• Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents [44.34340798542]
  Large Language Models (LLMs) have shown remarkable capabilities in natural language tasks requiring complex reasoning. Traditional supervised pre-training on static datasets falls short in enabling autonomous agent capabilities. We propose a framework that combines guided Monte Carlo Tree Search (MCTS) search with a self-critique mechanism and iterative fine-tuning on agent interactions.
  arXiv  Detail & Related papers  (2024-08-13T20:52:13Z)
• DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning [56.887047551101574]
  We present DS-Agent, a novel framework that harnesses large language models (LLMs) agent and case-based reasoning (CBR) In the development stage, DS-Agent follows the CBR framework to structure an automatic iteration pipeline, which can flexibly capitalize on the expert knowledge from Kaggle. In the deployment stage, DS-Agent implements a low-resource deployment stage with a simplified CBR paradigm, significantly reducing the demand on foundational capabilities of LLMs.
  arXiv  Detail & Related papers  (2024-02-27T12:26:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.