laguarage/bert-finetuned-squad
Question Answering
•
0.1B
•
Updated
•
4
Shanshan Zhu
laguarage
·
AI & ML interests
recommender system, OCR, text2img
Recent Activity
reacted
to
mike-ravkine's
post
with 👀
about 2 months ago
Spatial reasoning is a domain where LLMs struggle surprisingly hard.
A new paper, "Stuck in the Matrix: Probing Spatial Reasoning in Large Language Models" compares performance on a handful of spacial reasoning tasks and finds all SOTA LLMs breaking down and hallucinating their faces off when the grids get large.
https://arxiv.org/html/2510.20198v1
The word search task is especially revealing: notice the bias towards detecting "horizontal" while struggling with "vertical" - LLMs only understand simple, linear relationships.. add a stride for 2D and it's basically over.
liked
a model
5 months ago
HuggingFaceTB/SmolVLM-Instruct
liked
a model
about 1 year ago
Sharpaxis/Mental-Health-RoBERTa
Organizations
