Petr Motlicek

A discriminative adaptation that reads the final prompt token's hidden state through a classification head, producing a label in one forward pass without modifying the backbone, gives a controlled comparison of generative and discriminative inference in an otherwise identical speechLLM.

Geometric Latent Reasoning reduces the length of reasoning chains in LLMs, achieving correct answers with fewer steps by leveraging continuous trajectories in embedding space.

Open-weight models can now generate realistic, long-form doctor-patient conversations with corresponding SOAP notes, providing a valuable resource for training and evaluating long-context audio reasoning systems.