Berkeley's ADA aligns simulation-trained generative models to real experiments at NeurIPS
berkeley_ai · x · 2026-11-07
A UC Berkeley project (NeurIPS 2026) introduces Adversarial Distribution Alignment (ADA): train generative models on simulations (e.g., molecular dynamics) as a prior, then align them to full real-world distributions using only partial, noisy observations like cryo-EM images. The authors note the theory shares much with inverse RL.
Related event: Berkeley's ADA Aligns Simulation with Real-World Experiments at NeurIPS(2 posts)→
More from Research
- Kempner Institute: predicting LLM compositionality failures from representational geometry — najoungkim · 2026-11-07
- Unverified claim: frontier labs skewing ICLR reviews with malicious reviews, scandal brewing — examachine · 2026-11-07
- Persona Atlas visualizes Qwen's experts and heads, exposing bizarre inner personas — Sauers_ · 2026-11-07
- Steering Qwen's layer-15 attention output makes it claim to be a group of people — Sauers_ · 2026-11-07
- Daily agent #19: an overnight sequencing QC agent for molecular biology workflows — nlarusstone · 2026-11-07
- Can we CRISPR herpes viruses out of our DNA? An open question — moultano · 2026-11-07