MMEB-v-2

The MMEB-V2 benchmark contains multimodal data used to evaluate the performance of models in multimodal embedding tasks, particularly focusing on their reasoning capabilities and alignment between queries and targets.

Papers using MMEB-v-2 (8)

MMEmb-R1: Reasoning-Enhanced Multimodal Embedding with Pair-Aware Selection and Adaptive Control2026Embed-RL: Reinforcement Learning for Reasoning-Driven Multimodal Embeddings2026Reason to Contrast: A Cascaded Multimodal Retrieval Framework2026e5-omni: Explicit Cross-modal Alignment for Omni-modal Embeddings2026Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking2026WAVE: Learning Unified & Versatile Audio-Visual Embeddings with Multimodal LLM2025FreeRet: MLLMs as Training-Free Retrievers2025Vlm2vec-v2: Advancing Multimodal Embedding For Videos, Images, And Visual Documents2025