1RL Environments and the Hierarchy of Agentic Capabilities (opens in new tab)(surgehq.ai)4echen4mo ago0
2Explaining Reinforcement Learning with Human Feedback (RLHF) (opens in new tab)(surgehq.ai)11echen3y ago0
3Users report Google Calendar bug creating random, fake events (opens in new tab)(theverge.com)2echen3y ago0
4McDonald’s is testing its first robot restaurant with no human contact (opens in new tab)(twistedfood.co.uk)2echen3y ago0
6ChatGPT Crushes Google on Coding Queries, and Matches It at General Information (opens in new tab)(surgehq.ai)11echen3y ago1
7AI Red Teams for Adversarial Training: Making ChatGPT and LLMs More Robust (opens in new tab)(surgehq.ai)9echen3y ago0
8HellaSwag: 36% of this popular large language model benchmark contains errors (opens in new tab)(surgehq.ai)49echen3y ago8
9The Violence, Racism, & Sexism Uncaught by Twitter's Content Moderation Systems (opens in new tab)(surgehq.ai)3echen3y ago0
10Move Over, Google: The TikTokification of Next-Gen Search (opens in new tab)(surgehq.ai)13echen3y ago4
11Sci-Fi Reddit Community Bans AI-Art for Being 'Low Effort' Posting (opens in new tab)(vice.com)2echen3y ago0
12Stability AI, the startup behind Stable Diffusion, raises $101M (opens in new tab)(techcrunch.com)13echen3y ago2
13DALL·E vs. Imagen, and Evaluating Astral Codex Ten's Bet on AI Progress (opens in new tab)(surgehq.ai)13echen3y ago0
14The $250K Inverse Scaling Prize and Human-AI Alignment (opens in new tab)(surgehq.ai)11echen3y ago0
15The company that makes GIFs says they're 'cringe' and 'out of fashion' (opens in new tab)(businessinsider.com)11echen3y ago0
