Behavioral Localization 方法让工具调用智能体更高效学习软技能
Robust Soft Skill Adaptation via Behavioral Localization in Tool-Using Agents
教智能体学技能不用整条轨迹都学，只挑 10% 关键位置就够，成功率还能涨 6 个多点，做 Agent 训练的可以看看。
论文提出 Behavioral Localization 方法，用于给使用工具的智能体训练紧凑的软提示技能。该方法对比冻结智能体在有、无文本技能时的表现，定位出技能真正起作用的轨迹位置。在 SpreadsheetBench 上用冻结的 Qwen3.6-35B-A3B 实验显示，轨迹中前 10% 的位置就贡献了 96.12% 的正向技能收益。共享提示设定下，选择性监督将成功率从全轨迹的 30.36% 提升到 36.79%；任务专属设定下达到 57.38%，超过完整文本技能和 OPCD 基线。
Robust Soft Skill Adaptation via Behavioral Localization in Tool-Using Agents
Text skills can extend the capabilities of tool-using agents, but long skill descriptions increase computational costs. Soft prompts provide a compact alternative, yet learning from entire successful trajectories may also introduce unnecessary behavioral changes. We find that the useful effect of a text skill is highly concentrated in a small subset of trajectory positions. Based on this observation, we propose Behavioral Localization, which compares the frozen agent with and without the text skill to identify positions where the skill both increases the likelihood of successful tokens and meaningfully changes the next token distribution. We use these localized positions to supervise a short soft prompt. Experiments on SpreadsheetBench with a frozen Qwen3.6-35B-A3B agent show that the top 10\% of trajectory positions capture 96.12\% of the total positive skill gain. In the shared prompt setting, selective supervision improves success from 30.36\% with full trajectory to 36.79\%. Under task-specific conditions, using the top 10\% localized positions achieves 57.38\%, outperforming both the full text skill and the on-policy OPCD baseline. These results show that a small set of behaviorally relevant trajectory positions can provide effective supervision of soft skills.
