Tool use / function calling

ReTool

ReTool: Reinforcement Learning for Strategic Tool Use in LLMs

Superseded baseline#15 of 55 most-superseded · first seen Apr 15, 2025

Superseded — cited as a baseline and beaten by newer methods

1 papers critique it · 1 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites ReTool as a baseline.

While these works advance our understanding of exploration in language models, none simultaneously learn exploration policies that select among heterogeneous tools while rewarding both answer quality and trajectory diversity.
Step-wise Policy for Rare-tool Knowledge (SPaRK): Offline RL that Drives Diverse Tool Use in LLMs

Beaten on benchmarks

Head-to-head results where a newer method reports beating ReTool. Values are copied from the source paper's tables — verify against the cited paper.

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.