agentic-abstention

HERA 提出了一种通过共同演化 agent harness 与任务环境来提升 LLM 智能体在不可行任务上弃权能力的框架，将弃权准确率从 61.7% 提升至 83.3%，且可跨 19 个模型迁移，并发布 HERA-Bench 评测基准。

This paper defines agentic abstention, the problem of deciding when an LLM agent should stop acting under uncertainty, and evaluates it across web shopping, terminal environments, and question answering. It introduces convolve, a context engineering method that improves timely abstention without updating model parameters.