让智能体用切换LoRA适配器的方式，不遗忘旧技能还能解锁新能力。
Agents unlock new capabilities through Switching LoRA Adapters as a Tool (SLAaaT)
- 智能体在执行任务时可自主切换不同专用LoRA适配器。
- 解决率提升至4个最难任务，令牌消耗减少46.1倍。
- 适合需要多技能长期协作的复杂代理系统。
后训练可激发新能力并提升特定任务表现，但常导致其他领域出现灾难性遗忘。这对包含多阶段能力组合的长周期智能体轨迹构成挑战。本文提出让智能体通过切换专用LoRA适配器作为工具，避免这一权衡。我们在两个合成编码任务上验证该方法：任务逻辑简单但需专业化处理。结果表明，该方法使模型解决此前无法完成的问题；能自主切换并发现优于人工启发式策略的新策略；相较仅使用单一适配器，能力损耗降低最高达18倍。此外，该方法在任务解决能力（成功解决4个最难题目，子代理为0）和令牌使用效率（某些场景减少46.1倍）上均显著超越子代理方案。
原文摘要 · Abstract (English)
Post-training can unlock new capabilities and improve performance on specialized tasks, but sometimes at the cost of catastrophic forgetting in other domains. This poses a problem in long agent trajectories that compose different capabilities. We reject this tradeoff by giving an agent a tool to switch between specialized LoRA adapters mid-trace. To test its effectiveness, we compose two synthetic coding tasks that are logically simple but require specialization. We find that this allows the model to solve problems it previously could not, that the model is able to switch autonomously (and find a new strategy that beats our human heuristic baseline on one task), and that this incurs an up to an 18x reduction in capability tax compared to an agent using only one specialized adapter. Our approach also substantially outperforms spawning subagents in both task capabilities (solving 4 of our hardest tasks versus none) and token usage (46.1x fewer tokens in some scenarios).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。