TextCraft

TextCraft is a dataset used to evaluate the performance of multi-agent workflows in completing long-horizon tool-using tasks through the analysis of cyclic subtask graphs.

Papers using TextCraft (3)

Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade2026GATE: Graph-based Adaptive Tool Evolution Across Diverse Tasks2025ADaPT: As-Needed Decomposition and Planning with Language Models2023 · 2 cites