新闻
What Breaks Under Pruning in Smart Homes, and When? Evaluating LLM Degradation Across Architectures and Task Complexity
arXiv cs.AI · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- Study shows how pruning degrades LLM performance in smart-home tool calling, with dense models failing sharply but mixture-of-experts models tolerating more pruning.
- 为何重要
- Engineers deploying pruned LLMs for home automation or device control need to understand failure modes beyond overall accuracy metrics.
- 注意
- Results are specific to smart-home tool calling tasks; degradation patterns may differ significantly for other LLM applications and use cases.
- llm
- language model
- tool calling
- fine-tun
- mixture-of-experts
这条新闻背后的模式
- Progressive Rollout & Shadow Mode
- Context Editing & Tool-Result Clearing
- Agentic Context Engineering (Evolving Playbook)
每个模式都讲清楚技术如何运作、何时值得投入，以及在哪里会失效。
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。