1
The Illusion of Perfect LLM Code
五大主流LLM同场竞技，揭穿“完美代码”谎言——真实任务、执行与安全漏洞全曝光。
Article URL: https://ariya.io/2026/05/the-illusion-of-perfect-llm-code/ Comments URL: https://news.ycombinator.com/item?id=48360797 Points: 3 # Commen…
2
LLM INQUISITOR: Evaluating how AI models handle long, realistic tasks
开源项目LLM INQUISITOR提供真实场景下的LLM行为评估框架，专注长任务和实用性，而非基准测试。
Article URL: https://github.com/AssimilatedHuman/LLM-Inquisitor Comments URL: https://news.ycombinator.com/item?id=48207330 Points: 1 # Comments: 0
3
Measuring the performance of our models on real-world tasks
OpenAI推出GDPval评估，衡量模型在44个职业真实经济任务中的表现，为AI实际价值提供新标尺。
OpenAI introduces GDPval, a new evaluation that measures model performance on real-world economically valuable tasks across 44 occupations.