Looped Language Models Improve Compositional Tool Calling

Hugging Face Daily Papers Papers
looped-language-models compositional-tool-calling agentic-systems adaptive-inference recurrent-computation multi-step-tool-use

Summary

Looped language models enhance compositional tool calling by leveraging recurrent computation, improving accuracy on multi-step tasks while adaptive inference optimizes the balance between performance and compute cost. The study suggests these models are promising for reliable agentic systems.

Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question in compositional tool-calling settings, where models must coordinate multiple API calls, maintain intermediate state, and preserve dependencies across tool interactions. We evaluate native and retrofitted looped language models on API-Bank, BFCL, and NESTful, comparing looped and non-looped models trained under matched supervised fine-tuning recipes and varying recurrent depth at inference time. In controlled experiments, recurrent computation generally benefits compositional and dependency-aware tool use, while providing smaller and more model-dependent gains on isolated API invocation. Accuracy on multi-step tool use generally increases with recurrent depth; adaptive inference, however, achieves a more favorable compute-performance trade-off by allocating additional computation only when needed. Our results suggest that looped language models are a promising architecture for agentic systems that require reliable planning, coordination, and execution of compositional tool use workflows.
Original Article
View Cached Full Text

Cached at: 08/20/26, 04:03 AM

Paper page - Looped Language Models Improve Compositional Tool Calling

Source: https://huggingface.co/papers/2608.18171

Abstract

Looped language models improve compositional, multi-step tool use through recurrent computation, with adaptive inference balancing accuracy and compute cost.

Looped language modelshave shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question incompositional tool-callingsettings, where models must coordinate multiple API calls, maintain intermediate state, and preserve dependencies across tool interactions. We evaluate native and retrofittedlooped language modelsonAPI-Bank,BFCL, andNESTful, comparing looped and non-looped models trained under matched supervised fine-tuning recipes and varyingrecurrent depthat inference time. In controlled experiments, recurrent computation generally benefits compositional and dependency-aware tool use, while providing smaller and more model-dependent gains on isolated API invocation. Accuracy onmulti-step tool usegenerally increases withrecurrent depth;adaptive inference, however, achieves a more favorable compute-performance trade-off by allocating additional computation only when needed. Our results suggest thatlooped language modelsare a promising architecture foragentic systemsthat require reliable planning, coordination, and execution of compositional tool use workflows.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2608\.18171

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.18171 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.18171 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.18171 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Allocating Recurrent Compute in Looped Language Models

arXiv cs.LG

This paper introduces MixerLoop, a method that allocates recurrent compute by selectively looping the mixer component in language models while applying the feed-forward network once, achieving performance improvements with reduced computational costs.

Looped State-Space Language Models with Adaptive Exit-State Selection

arXiv cs.AI

This paper explores looped (recurrent) state-space language models using Mamba and hybrid Mamba-Transformer backbones, showing they outperform non-looped baselines on reasoning tasks and remain competitive under iso-parameter and iso-FLOPs pretraining, with adaptive exit-state selection improving intermediate-depth performance.

Recursive Language Models

Papers with Code Trending

This paper introduces Recursive Language Models (RLMs), an inference strategy that enables LLMs to process arbitrarily long prompts by treating them as external environments and recursively calling themselves over prompt snippets. RLMs handle inputs two orders of magnitude beyond context windows and outperform base LLMs on long-context tasks with comparable cost.

Looped World Models

Hugging Face Daily Papers

Looped World Models introduce iterative latent state refinement through shared transformer blocks, achieving 100x parameter efficiency while adapting computational depth to prediction complexity.