[AINews] Lilian Weng summarizes 35 papers on Harness Engineering for RSI a quiet day lets us read some condensed insight Congrats to Meta Superintelligence on having the top 2/3 image/video models in the world! This would’ve been a candidate for a title story, but unfortunately that is pretty much all the detail we have about Muse Image/Video - no paper, no technical detail whatsoever. Still, this beats the Microsoft MAI models from last month which is nice. We are noted Lilian Weng fans, so we take notice whenever she drops another research recap, especially rare now that she is a cofounder a