3 papers
cs.CL2024
Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity
Rheeya Uppaal, Apratim Dey, Yiting He +2
Recent alignment algorithms such as direct preference optimization (DPO) have been developed to improve the safety of large language models (LLMs) by training these models to match…
math.AP2021
Stable solution of the log-Minkowski problem in the case of many hyperplane symmetries
Karoly J. Boroczky, Apratim De
In the case of symmetries with respect to n independent linear hyperplanes, the stability of the solution of the Logarithmic Minkowski problem on S^{n-1} is established.
math.CA2020
Stability of the Prekopa-Leindler inequality for log-concave functions
Karoly J. Boroczky, Apratim De
Stability version of the Prekopa-Leindler inequality for log-concave functions on the n-dimensional Euclidean space is established.