Every single review I have handled at NeurIPS as an AC is **much worse** than a well executed AI review. By "much worse" I do not mean 20% or 50%. I mean a factor of about 10-100. At least - 10x more mathematical issues are caught, - 10x more missing key citations are caught, - 10x more novelty claims are invalidated with evidence, - 10x more issues with the experiments are observed, - 10x more typos and grammatical issues are fixed, - 10x more inconsistencies are found, and so on. In many cases I have handled over the last 10 years, the human reviews are so bad in comparison that the improvement factor is closer to 100. (The best human reviews I received over the years are worse than an AI review I can generate today, but still entirely good enough to make the correct decision. On the other hand, an AI review is often an almost comprehensive summary of all key issues --- something a human almost never has time to deliver --- and such feedback is immensely useful to the authors) At this point, we would be much better to just 1) give one very thorough ChatGPT review to all submitted papers (I am mainly talking about my fields - optimization, AI, machine learning), automatically, and 2) keep asking for a revision or two (to keep the duration of the process within some bounds) until the number of issues decreases to an extent when the AI reviewer is satisfied, in a given time-frame (eg, 1 month). 3) At that point, a human AC can make a decision (to keep an eye on this all should anything go wrong). The role of the AC would be merely to observe and manage the process, and make a final decision, based on the trajectory of the revisions. I can't believe I am saying this -- AI reviews were a nonsense idea even a year ago. The current AI reviews are super-human.

Introducing H3 Max, new post-trained video model by fal Research. H3 Max ranks #1# for overall quality, prompt understanding, and aesthetics against leading video models, on both first-party and third-party independent evaluations while generating a 5-second 720p video under 3 seconds. H3 Max is 50% off for the next week, making it the highest quality, fastest and cheapest model for overall video generation tasks.

@rperezmarco I think this is somewhat standard practice around old mathematicians, for better or worse - no one denounced hironaka’s claimed resolution of singularities in char p, but also no one took it particularly seriously

AheadForm just raised a new A1 round worth hundreds of millions of RMB (~tens of millions USD) 🤖 That matters because this is not another humanoid company chasing locomotion first. AheadForm is building around the part most robotics startups still underestimate: face, emotion, and real-time human connection. The new funding will go into multimodal embodied interaction, emotion foundation models, facial hardware and materials, standardized delivery, and global expansion. Founded in June 2024, the company is still young, but the founder’s research trail is not. Yuhang Hu, a Columbia PhD and AheadForm’s CEO/CTO, has published work spanning facial coexpression, realistic lip motion for humanoid face robots, and self-supervised robot self-modeling. That is the deeper signal here. In a market crowded with hands, arms, and walking demos, investors are now putting serious money behind embodied AI that can express, respond, and hold attention face to face. And the company is moving fast. According to public reports, AheadForm has completed five funding rounds since the second half of 2025, while its robots have already broken out of lab-only visibility through public activations like the NetEase Justice mobile game collaboration and large robot-stage appearances. If humanoid robotics is about physical labor, AheadForm is making the case that the next layer is emotional presence. That may end up being one of the more important categories in embodied AI.