Vincent Counathe

Quantization and low-precision training for large language models.

I am pursuing a PhD at Cornell University, advised by Chris De Sa, and work on quantization at Together AI.

Selected Research

QUASAR: Lowering the Loss Floor of Quantization-Aware Training with Loss-Aware Reconstruction Vincent Counathe, Ben Athiwaratkun, Christopher De Sa, Tianyi Zhang
arXiv preprint, 2026

QUASAR treats quantization-aware training as a weight reconstruction problem, continuously fitting a loss-aware quantized representation as the latent full-precision weights evolve. Its analysis connects weight reconstruction quality to the optimization dynamics and final quantized-model loss.

Paper · Code · Checkpoints · Technical note (forthcoming)

Background

Before Cornell, I completed my Master's in Statistics and Machine Learning at Université Paris-Saclay (Institut de Mathématiques d'Orsay), where I worked on high-dimensional statistics and machine learning with Florent Krzakala, Lenka Zdeborová, and Christophe Giraud.

Earlier Research in Probability and Statistics

Statistical Risk Bounds for Genealogical Reconstruction on Random Recursive Trees Vincent Counathe

Non-asymptotic risk bounds for a Jordan-ordering estimator in genealogical inference. [Paper]

Localization Schemes for Mixing Bounds in Markov Chains Vincent Counathe, Matthieu Dagès

Expository note on the Chen–Eldan framework. [Note]

Community Detection in Random Graphs Vincent Counathe, Matthieu Dagès

Expository note on spectral methods for community detection in random graphs. [Note]