Lattice AI Research

Research focus

Natural Language Processing (2)Training Efficiency & Optimization (2)Computer Vision (1)Architecture Design (Transformers, SSMs, MoE) (1)

Frequent co-authors

Mehryar Mohri (2)Yutao Zhong (2)Corinna Cortes (1)Anqi Mao (1)

Papers (2)

Apr 30, 2026

Corinna Cortes +43w ago

Optimized Deferral for Imbalanced Settings

Expert imbalance can cripple learning-to-defer systems, but a novel cost-sensitive margin-based loss function can restore performance.

Corinna Cortes, Anqi Mao, M. Mohri +2

Computer Vision Natural Language Processing Training Efficiency & Optimization

M. Mohri +23w ago

Linear-Core Surrogates: Smooth Loss Functions with Linear Rates for Classification and Structured Prediction

Get the best of both worlds: Linear-Core Surrogates offer the fast optimization of smooth losses and the statistical efficiency of margin-based losses, without sacrificing differentiability.

M. Mohri, Mehryar Mohri, Yutao Zhong

Architecture Design (Transformers, SSMs, MoE)Natural Language Processing Training Efficiency & Optimization

Search

M. Mohri

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)