DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents

One-line takeaway

Google's DeLoop tackles loops in long-horizon agents — re-issuing earlier tool calls that return identical observations — seen natively in Qwen, Llama and Gemini agents and worse under history compression; it self-supervises from the agent's own rollouts, using looping actions as negative examples.

Google 的 DeLoop 处理长任务智能体“兜圈子”：重复发出已经调用过、结果完全相同的工具调用。这种循环在 Qwen、Llama、Gemini 智能体里都会出现，历史被压缩后更常见；DeLoop 用智能体自己的轨迹做自监督，把循环动作当负例来训练。