André Freitas

Structured proxy judgments can refine reasoning outputs without needing exact gold standards, leading to substantial improvements in performance.

Code isn't just for computation: it's a better latent representation of mathematical concepts than pure visuals, and can be used to boost geometric reasoning in MLLMs.

LLMs encode *how* a hypothesis relates to a premise in NLI, not just *that* it relates, opening new avenues for mechanistic interpretability.

Turns out, language models can reason about mechanical engineering problems, iteratively refining linkage designs by diagnosing failure modes and proposing grounded corrections, all without fine-tuning.