【AI资讯】6月20日
【AI资讯】6月20日
本文介绍了Ilya Sutskever成立的新公司SSI，专注于构建安全超级智能；北京大学推出的机器人多模态大模型RoboMamba，具备出色的推理能力；上海交通大学生成式人工智能实验室推出的大模型认知推理能力评估基准OlympicArena；以及GPT-4o在ARC-AGI基准测试中的突破性表现。
10大AI搜索终极比拼，谁最强？超详细评测：Perplexity、秘塔、百小应、360AI、天工、Thinkany、GPT-4o等
10大AI搜索终极比拼，谁最强？超详细评测：Perplexity、秘塔、百小应、360AI、天工、Thinkany、GPT-4o等
This document discusses the comparison of 10 AI search tools and evaluates their performance in answering specific questions. The evaluation covers various aspects such as the accuracy and completeness of the answers provided by each tool. The results show that 百小应 performed the best, followed by Perplexity and GPT. The document also highlights some limitations of AI search tools, including their inability to recognize internet memes, susceptibility to generating misleading information, and difficulty in handling conflicting information.