InfiAgent: An Infinite-Horizon Framework for General-Purpose Autonomous Agents
Chenglin Yu, Yuchen Wang, Songmiao Wang, Hongxia Yang, Ming Li

TL;DR
InfiAgent introduces a framework for autonomous agents that maintains a bounded reasoning context over indefinite periods by externalizing state, enabling stable long-horizon reasoning without task-specific fine-tuning.
Contribution
It proposes a novel externalized state management approach that prevents context growth, improving long-term reasoning stability in autonomous agents.
Findings
InfiAgent with a 20B model performs competitively with larger proprietary systems.
Maintains higher long-horizon coverage than context-centric baselines.
Effective without task-specific fine-tuning.
Abstract
LLM agents can reason and use tools, but they often break down on long-horizon tasks due to unbounded context growth and accumulated errors. Common remedies such as context compression or retrieval-augmented prompting introduce trade-offs between information fidelity and reasoning stability. We present InfiAgent, a general-purpose framework that keeps the agent's reasoning context strictly bounded regardless of task duration by externalizing persistent state into a file-centric state abstraction. At each step, the agent reconstructs context from a workspace state snapshot plus a fixed window of recent actions. Experiments on DeepResearch and an 80-paper literature review task show that, without task-specific fine-tuning, InfiAgent with a 20B open-source model is competitive with larger proprietary systems and maintains substantially higher long-horizon coverage than context-centric…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsExplainable Artificial Intelligence (XAI) · Advanced Software Engineering Methodologies · Context-Aware Activity Recognition Systems
