Azalia Mirhoseini

Cartridges: Lightweight and general-purpose long context representations via self-study

Support Your Local LMs: Redistributing LM Traffic from Cloud to Edge with TrafficBench

On the Role of Temperature Sampling in Test-Time Scaling

How Do Large Language Monkeys Get Their Power (Laws)?

SPRINT: Enabling Interleaved Planning and Parallelized Execution in Reasoning Models

Exploring Diffusion Transformer Designs via Grafting

Synthetic Data Generation and Multi-Step Reinforcement Learning for Reasoning and Tool Use

Reasoning-SQL: Reinforcement Learning with SQL Tailored Partial Rewards for Reasoning-Enhanced Text-to-SQL

Weaver: Shrinking the Generation-Verification Gap by Scaling Compute for Verification

KernelBench: Can LLMs Write Efficient GPU Kernels?

Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

Archon: An Architecture Search Framework for Inference-Time Techniques

An Architecture Search Framework for Inference-Time Techniques