Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoning Paper • 2607.19790 • Published 3 days ago • 4
SPHINX Models Collection Qwen3 model checkpoints trained on SPHINX https://huggingface.co/datasets/maveryn/sphinx. • 3 items • Updated 10 days ago • 1
Beyond Perception Errors: Semantic Fixation in Large Vision-Language Models Paper • 2604.12119 • Published Apr 13 • 2