Our call for papers is live!! The 7th Workshop on Gender Bias in NLP will be co-located with AACL and will be remote-only. Deadline for direct submissions is September 2.
What Makes Linguistic Representations Good Models of High-Level Visual Perception in the Human Brain?
This question underlies the study presented in our new preprint. We trained brain-encoding models to predict fMRI responses to naturalistic images from LLM-derived embeddings of image captions.
In our paper, we show that conversation topics are the main predictors of LLM-generated high-stakes advice within a conversational context and can act as proxies for sociodemographics. We also demonstrate that LLMs actually struggle to infer user sociodemographics from a conversation history.
š2nd Call for Papers!š EMNLP 2026 Workshop on Multimodal Interaction in Face-to-Face Dialogue (MINT) welcomes work on multimodal communication, speech, gesture, gaze, facial expressions, datasets, eval methods, and applications! @emnlpmeeting.bsky.social mintworkshop.github.io