Last week, I presented my recent paper on memorization and hallucination in LVLMs at
@l2m2_workshop #ACL2025NLP.
TLDR: model memorization may be more prevalent over image sources vs. text! Details here: aclanthology.org/2025.l2m2-1.10/
Peter Carragher
@PeterVCarragher
Societal Computing PhD student @S3DatCMU
- Just wrapped up my first IC2S2 conference with a talk on our ongoing research into how broadcasting companies influence editorial decisions in news outlets. Check out our News Guessing game at newsguess.petercarragher.com and contribute to our research on media ownership!
- Excited to have two workshop papers and one main conference paper that I've been involved in being presented at @icwsm! Thanks @CatherineKing77 and @ZupEvan for letting me tag along. Details below.
- Preprint! Vision-language models like GPT-4o fail to identify multimodal knowledge conflicts and frequently hallucinate on counterfactual samples. Try out the batsman example below (promt = "what is he holding?").
- Our recent paper on source credibility and link spam in Webgraphs got featured on the @CommonCrawl site after being included earlier this year in a special issue on responsible recommender systems in TIST. Check it out here: dl.acm.org/doi/full/10.11… @ZupEvan
