Web Science and Digital Libraries Research Group

Image Generated by DALL-E The utility of Large Language Models (LLMs) across diverse tasks raises important questions about the role of AI in web archive contexts, particularly in facilitating change analysis tasks. For instance, can LLMs effectively detect and analyze changes in webpages? How can this be achieved? What advantages and limitations do AI approaches have compared to traditional methods?.  In this blog post, I summarize our paper  “Exploring Large Language Models for Analyzing Changes in Web Archive Content: A Retrieval-Augmented Generation Approach”  ( Jhon G. Botello ,  Lesley Frew ,  Jose J. Padilla ,  Michele C. Weigle ) published at the  9th Computational Archival Science (CAS) Workshop 2024  (check Lesley Frew's trip report for more details about the workshop). The paper represents our first step in determining the ability of LLMs and the extent to which they can assist with change analysis tasks. Motivation As we interact...

Semantic mapping of changes between images using Triplet Loss Metric Learning, Fig 8. from Zhang et al . I talked about two kinds of trust in my previous two posts,  Evaluating Trust in User-Data Networks: What Can We Learn from Waze?  and Trust Management in Multi-Agent Systems via Deep Reinforcement Learning . In the former, we looked at trust as a measure of the accuracy of data provided by user and in the latter we looked at evaluating the behavior of the user to measure out trust in that particular user. The difference is subtle but apparent - a user with trustworthy historical behavior consistently provides accurate data. Conversely, one who provides data with inconsistent accuracy can be considered as being a qualitatively inferior data provider with measurably lower trust.  This implies an exploitable attack vector, however. If we award implicit trust based on historical behavior, what happens when a historically trustworthy user suddenly provides the system with ...