Megadose The AI curve, compressed.

JET: Just-Enough Exploration Tuning for Efficient Tool-Calling Agents in Repository-Level Question Answering

· ArXiv · AI/CL/LG ·
JET cut tool calls by 11–33% in repository QA tests while largely preserving answer scores.

The preprint finds that agents keep exploring code after collecting enough information to answer, and supervised fine-tuning can pass that behavior on. JET cuts training trajectories at the point where sufficient information is available, adds a rationale for answering there, and gives that rationale extra weight during training. Tests covered two benchmarks and two open-weight 8B models. With expert trajectories, JET used 18–28% fewer tool calls than standard supervised fine-tuning while retaining most of its score gains; inference costs fell in both settings.

ArXiv · AI/CL/LG's note

score 5

Categories: Research