GOAT-Bench

The GOAT Benchmark (HomePage) We introduce the GOAT-Bench, a comprehensive and specialized dataset designed to evaluate large multimodal models through meme-based multimodal social abuse. GOAT-Bench comprises over 6K diverse memes, encompassing a range of themes including hate speech and offensive content. Our focus is

Papers using GOAT-Bench (7)

RoboAtlas: Contextual Active SLAM2026GoalVLM: VLM-driven Object Goal Navigation for Multi-Agent System2026LangMap: A Hierarchical Benchmark for Open-Vocabulary Goal Navigation2026HIMM: Human-Inspired Long-Term Memory Modeling for Embodied Exploration and Question Answering2026AstraNav-Memory: Contexts Compression for Long Memory2025MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation2025LagMemo: Language 3D Gaussian Splatting Memory for Multi-modal Open-vocabulary Multi-goal Visual Navigation2025