Temporal Alignment of Time Sensitive Facts with Activation Engineering
This addresses the issue of temporal relevance in LLM outputs for users needing accurate, time-sensitive information, representing an incremental improvement in efficiency over existing methods.
The paper tackles the problem of ensuring large language models generate time-appropriate factual responses by using activation engineering to temporally align models without training or dataset creation, achieving up to 44% improvement in relative prompting and comparable performance to fine-tuning methods.
Large Language Models (LLMs) are trained on diverse and often conflicting knowledge spanning multiple domains and time periods. Some of this knowledge is only valid within specific temporal contexts, such as answering the question, "Who is the President of the United States in 2022?" Ensuring LLMs generate time appropriate responses is crucial for maintaining relevance and accuracy. In this work we explore activation engineering as a method for temporally aligning LLMs to improve factual recall without any training or dataset creation. In this research we explore an activation engineering technique to ground three versions of LLaMA 2 to specific points in time and examine the effects of varying injection layers and prompting strategies. Our experiments demonstrate up to a 44% and 16% improvement in relative and explicit prompting respectively, achieving comparable performance to the fine-tuning method proposed by Zhao et al. (2024) . Notably, our approach achieves similar results to the fine-tuning baseline while being significantly more computationally efficient and requiring no pre-aligned datasets.