斯坦福與 DeepMind 的新論文指出,可藉由演化現有工具使用任務,使測試更具挑戰性但仍可解決。
\n- 演化方法:使用 TURTLE 搭配 AlphaEvolve,對 AppWorld 任務加入濾鏡、額外步驟與隱藏查詢,並以模型跑10次,目標成功率約 50%。
AI 伴隨問答
針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:
New Stanford and Google DeepMind paper shows that you can turn too-easy tool-use benchmarks into hard, still-solvable tests by evolving existing tasks against the model.
Tool-use benchmarks like AppWorld are static, so models soon solve most tasks, and building new ones takes human work.
TURTLE uses AlphaEvolve to mutate existing tasks, adding filters, extra steps and hidden lookups, then runs the solver model 10 times on each. Tasks score highest when the solver succeeds about 50% of the time, and each must be solved at least once.
69% of original AppWorld tasks were always solved by Gemma 4 31B-it, while nearly all 735 evolved tasks sat at 40% to 60% completion. An LLM judge rated 52.5% of evolved tasks hard, versus 19.8% of the originals.
If your model aces a benchmark, evolve its tasks toward a 50% success rate instead of writing a new benchmark from scratch.