9 papers
Model Directions, Not Words: Mechanistic Topic Models Using Sparse Autoencoders
Carolina Zheng, Nicolas Beltran-Velez, Sweta Karlekar +5
Traditional topic models are effective at uncovering latent themes in large text collections. However, due to their reliance on bag-of-words representations, they struggle to captu…
The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL
Nicolas Beltran-Velez, Felix Friedrich, Zhang Xiaofeng +4
Score- and flow-matching models often rely on preference-based reinforcement learning for two purposes: aligning with subjective preferences and, surprisingly, recovering propertie…
Inference-time Physics Alignment of Video Generative Models with Latent World Models
Jianhao Yuan, Xiaofeng Zhang, Felix Friedrich +7
State-of-the-art video generative models produce promising visual content yet often violate basic physics principles, limiting their utility. While some attribute this deficiency t…
Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences
Sweta Karlekar, Carolina Zheng, Magnus Saebo +5
Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space. Existing methods use a calibrat…
Improving the Physics of Video Generation with VJEPA-2 Reward Signal
Jianhao Yuan, Xiaofeng Zhang, Felix Friedrich +7
This is a short technical report describing the winning entry of the PhysicsIQ Challenge, presented at the Perception Test Workshop at ICCV 2025. State-of-the-art video generative…
Estimating the Hallucination Rate of Generative AI
Andrew Jesson, Nicolas Beltran-Velez, Quentin Chu +5
This paper presents a method for estimating the hallucination rate for in-context learning (ICL) with generative AI. In ICL, a conditional generative model (CGM) is prompted with a…