← Search

Tianren Peng

2 accepted papers

2026

LoRA-DA: Data-Aware Initialization for Low-Rank Adaptation via Asymptotic Analysis

ICML 2026poster

LoRA has become a widely adopted method for PEFT, and its initialization methods have attracted increasing attention. However, existing methods have notable limitations: many methods do not incorporate target-domain data, while gradient-based methods exploit data only at a shallow level by relying o…

Cited by 0SourceScholar
2025

A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning

NeurIPS 2025poster

Multi-source transfer learning provides an effective solution to data scarcity in real-world supervised learning scenarios by leveraging multiple source tasks. In this field, existing works typically use all available samples from sources in training, which constrains their training efficiency and m…

Cited by 0SourcecodeScholar