← all datasets

MIA

Emerging
1papers using it
2025first seen

The 'MIA' dataset/benchmark is a challenging and diverse multi-modal instruction-following evaluation tool that assesses how well Multi-modal Large Language Models (MLLMs) understand and execute user instructions, incorporating both output response constraints and input image perception constraints.

Papers using MIA (1)

MIA dataset β€” papers, benchmarks & downloads Β· Large Language Models