← Search

Anish Kachinthaya

2 accepted papers

2025

Interpreting and Editing Vision-Language Representations to Mitigate Hallucinations

ICLR 2025poster

We investigate the internal representations of vision-language models (VLMs) to address hallucinations, a persistent challenge despite advances in model size and training. We project VLMs’ internal image representations to their language vocabulary and observe more confident output probabilities on…

2024

ALOHa: A New Measure for Hallucination in Captioning Models

NAACL 2024short

Despite recent advances in multimodal pre-training for visual description, state-of-the-art models still produce captions containing errors, such as hallucinating objects not present in a scene. The existing prominent metric for object hallucination, CHAIR, is limited to a fixed set of MS COCO objec…

Cited by 12SourcePDFScholar