Enhancing Learned Knowledge in LoRA Adapters Through Efficient Contrastive Decoding on Ascend NPUs
Open Access
- 3 August 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 4499-4510
Abstract
No abstract availableAll Related Versions
This publication has 17 references indexed in Scilit:
- Enhancing Recommendation Diversity by Re-ranking with Large Language ModelsACM Transactions on Recommender Systems, 2025
- RareBench: Can LLMs Serve as Rare Diseases Specialists?2024
- Xinyu: An Efficient LLM-based System for Commentary Generation2024
- Know Your Needs Better: Towards Structured Understanding of Marketer Demands with Analogical Reasoning Augmented LLMs2024
- On Adaptive Knowledge Distillation with Generalized KL-Divergence Loss for Ranking Model Refinement2024
- A Thorough Examination of Decoding Methods in the Era of LLMs2024
- Efficient Memory Management for Large Language Model Serving with PagedAttention2023
- AdapterSoup: Weight Averaging to Improve Generalization of Pretrained Language Models2023
- BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models2022
- AdapterHub: A Framework for Adapting Transformers2020