IAAILooped LMs for Tool Calling
Looped Language Models Improve Compositional Tool Calling
Evaluates native and retrofitted looped language models on compositional tool use, where models must coordinate several API calls and keep dependencies, comparing looped and non-looped models under matched fine-tuning while varying recurrent depth at inference.
Inside the method
[ Shared layers ] × r → multi-step API calls
Simplified conceptual schematic. Consult the paper for the complete architecture.
- Recurrence family
- Full-stack recurrence
- Depth control
- Varied inference depth; adaptive inference compared
- KV / state strategy
- See paper
Reading note
Recurrence mostly helps compositional, dependency-aware tool use on API-Bank, BFCL and NESTful, with smaller gains on single API calls. Adaptive inference gives a better compute-performance trade-off than fixed deep recurrence.
Sources checked 2026-10-28. This catalog does not imply independent reproduction.
Related works
Cite this work
@misc{loopedtoolcalling2026,
title = {Looped Language Models Improve Compositional Tool Calling},
author = {Andrei Cristian Popescu and Haitz Sáez de Ocáriz Borde and Pietro Liò},
year = {2026},
eprint = {2608.18171},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.18171}
}