Posts by Collection

portfolio

publications

The Labyrinth of Links: Navigating the Associative Maze of Multi-modal LLMs

Published in ICLR 2025, 2024

In this paper, we introduce a benchmark to evaluate association, a core yet underexplored aspect of human intelligence. We construct adjective- and verb-based tasks using an annotation-free method and propose three association levels: single-step, synchronous, and asynchronous. Extensive zero-shot evaluations reveal that current MLLMs, including GPT-4V, perform significantly worse than humans.

Download Paper

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.