Rethinking the Role of Tool Graph Memory in Multi-Turn Tool Use
Large language model (LLM) agents solve complex tasks by calling external tools over multiple turns. Recent work has proposed tool graph memory, which stores past experience as tool transitions to guide subsequent selection. However, it remains unclear whether this component consistently improves agents' tool use or undermines it. We evaluate tool graph memory on nine models and two benchmarks,…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
