collaborators

7 papers

cs.LG2025

In Good GRACEs: Principled Teacher Selection for Knowledge Distillation

Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi +2

Knowledge distillation is an efficient strategy to use data generated by large "teacher" language models to train smaller capable "student" models, but selecting the optimal teache…

stat.ML2025

The Coverage Principle: How Pre-Training Enables Post-Training

Fan Chen, Audrey Huang, Noah Golowich +5

Language models demonstrate remarkable abilities when pre-trained on large text corpora and fine-tuned for specific tasks, but how and why pre-training shapes the success of the fi…

astro-ph.IM2025

AstroMMBench: A Benchmark for Evaluating Multimodal Large Language Models Capabilities in Astronomy

Jinghang Shi, Xiaoyu Tang, Yang Huang +4

Astronomical image interpretation presents a significant challenge for applying multimodal large language models (MLLMs) to specialized scientific tasks. Existing benchmarks focus…

cs.CL2025

Metadata Conditioning Accelerates Language Model Pre-training

Tianyu Gao, Alexander Wettig, Luxi He +3

The vast diversity of styles, domains, and quality levels present in language model pre-training corpora is essential in developing general model capabilities, but efficiently lear…

cs.LG2025

Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization

Noam Razin, Sadhika Malladi, Adithya Bhaskar +3

Direct Preference Optimization (DPO) and its variants are increasingly used for aligning language models with human preferences. Although these methods are designed to teach a mode…

cs.LG2025

Provable unlearning in topic modeling and downstream tasks

Stanley Wei, Sadhika Malladi, Sanjeev Arora +1

Machine unlearning algorithms are increasingly important as legal concerns arise around the provenance of training data, but verifying the success of unlearning is often difficult.…