sujay sanghavi

Associate Professor@University of Texas, Austin·美国·OpenReview

研究方向

optimization · robustness · sparsity · high dimensional statistics · active learning · theory · language models · transformer architecture · parameter efficient fine tuning · distillation

5.0

SFT Doesn’t Always Hurt General Capabilities: Revisiting Domain-Specific Fine-Tuning in LLMs

合作者 (20)

Atula Tejaswi Neerkaje

sujay sanghavi

SFT Doesn’t Always Hurt General Capabilities: Revisiting Domain-Specific Fine-Tuning in LLMs

HiSpec: Hierarchical Speculative Decoding for LLMs

MuSED-FM: A Benchmark for Evaluating Multivariate Time Series Foundation Models

Sculpting Latent Spaces With MMD: Disentanglement With Programmable Priors

Enabling Approximate Joint Sampling in Diffusion LMs

Context-Free Synthetic Data Mitigates Forgetting

Understanding Contrastive Learning via Gaussian Mixture Models

RARe: Retrieval Augmented Retrieval with In-Context Examples