过度安全护栏让Claude连基础免疫学问题都频繁误判，生物学研究者直言工具已基本不可用。
I've found that doing basic biology research with Claude has become extremely difficult. I'm not talking about anything remotely related to developing…
恢复Hacker News隐藏功能，一键屏蔽不感兴趣帖子，支持自定义过滤规则，提升浏览体验。
The hide link seems to have disappeared from index pages. It’s still there on posts, but it doesn’t actually hide the post? I regularly used the hide …
用后备链拯救大模型过度拒绝问题，实测2%-8%误封可恢复。
Every production LLM app eats false-positive refusals. A user asks something perfectly fine, the safety filter trips, the model emits two sentences of…
高效多任务安全分类器，低成本实时检测毒性、越狱、仇恨言论等，专为LLM安全防护设计。
arXiv:2605.29659v1 Announce Type: cross Abstract: Real-time safety filtering for large language model (LLM) applications requires classifiers that can…