2 papers
cs.CL2024
Would I Lie To You? Inference Time Alignment of Language Models using Direct Preference Heads
Avelina Asada Hadji-Kyriacou, Ognjen Arandjelovic
Pre-trained Language Models (LMs) exhibit strong zero-shot and in-context learning capabilities; however, their behaviors are often difficult to control. By utilizing Reinforcement…
cs.LG2023
Context-PEFT: Efficient Multi-Modal, Multi-Task Fine-Tuning
Avelina Asada Hadji-Kyriacou, Ognjen Arandjelovic
This paper introduces a novel Parameter-Efficient Fine-Tuning (PEFT) framework for multi-modal, multi-task transfer learning with pre-trained language models. PEFT techniques such…