🏢 ETH Zurich

Relightable Full-Body Gaussian Codec Avatars: Realistic, animatable full-body avatars are now possible using learned radiance transfer and efficient 3D Gaussian splatting.

Democratizing advanced reasoning in AI, this blueprint introduces a modular framework for building Reasoning Language Models (RLMs), simplifying development and enhancing accessibility.

GSTAR: A novel method achieving photorealistic rendering, accurate reconstruction, and reliable 3D tracking of dynamic scenes with changing topology, even handling surfaces appearing, disappearing, or…

UIP2P: Unsupervised instruction-based image editing achieves high-fidelity edits by enforcing Cycle Edit Consistency, eliminating the need for ground-truth data.

LoRACLR merges multiple LoRA models for high-fidelity multi-concept image generation, using a contrastive objective to ensure concept distinctiveness and prevent interference.

LLMs’ hallucinations stem from entity recognition: SAEs reveal model ‘self-knowledge’, causally affecting whether it hallucinates or refuses to answer. This mechanism is even repurposed by chat finet…