AI Paper Atlas

Research map / Large language models / Agents, code and math

GTool: Graph Enhanced Tool Planning with Large Language Model

Wenjie Chen, Di Yao, Wenbin Li, Xuying Meng, Chang Gong, Jingping Bi

ICLR 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add request-specific node to GTool (ablation)GTool · HuggingFacen-F1: 0.7128 -> 0.7913Improved
add missing dependency prediction (MDPL) to GTool (ablation)GTool · HuggingFacen-F1: 0.7650 -> 0.7913Improved
add missing dependency prediction (MDPL) to GTool (ablation)GTool · HuggingFaceNED: 0.2541 -> 0.2537No clear effect
add both request-specific node and MDPL to GTool (ablation)GTool · HuggingFacen-F1: 0.6131 -> 0.7913Improved
add tool graph token to LLM prompt (ablation)GTool · HuggingFacen-F1: 0.1566 -> 0.7913Improved
add request-specific node on Vicuna-13B (ablation)GTool · HuggingFacen-F1: 0.7589 -> 0.8029Improved
add MDPL on Vicuna-13B (ablation)GTool · HuggingFacen-F1: 0.7707 -> 0.8029Improved
use GTool graph token instead of GNN4PlanGNN4Plan · HuggingFacen-F1: 0.4853 -> 0.7913Improved
use GTool instead of HuggingGPT on ToolE with Vicuna-13BHuggingGPT · ToolEl-F1: 0.3740 -> 0.3500Hurt
use GTool instead of HuggingGPT on ToolBenchHuggingGPT · ToolBenchn-F1: 0.5989 -> 0.6126Improved
increase MDPL sample size alpha from 4 to 6 (ablation)GTool · TaskBenchn-F1: 0.13% gainNo clear effect
increase loss weight lambda beyond 0.1 (ablation)GTool · TaskBenchperformance: peaks at lambda=0.1 then degradesMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy