EPIC-Bench

🎯 EPIC-Bench: A Perception-Centric Benchmark for Fine-Grained Embodied Visual Grounding in Vision-Language Models 📃 Overview 📚 EPIC-Bench is a Mask-Grounding-based benchmark designed to evaluate a VLM’s Visual Perception capability in Embodied Scenarios. EPIC-Bench covers 3 High-Level Categories and 23 Task Types, following the realistic Embodied Workflow: 🎯 TargetLocalization: Pinpoint the right object in the scene from a natural-language instruction. 🧭… See the full description on the dataset page:

Source: Hugging Face Hub (rxc205/EPIC-Bench). Metadata imported from the dataset’s Hub tags.