EmbeddingGemma 2 提出統一且高效的多模態表示
EmbeddingGemma 2: Unified and Efficient Multimodal Representations
EmbeddingGemma 2 是一款開放的多模態嵌入模型,在統一空間表示文字、圖像、音訊、影片及其組合。在涵蓋各模態的嵌入基準中,模型建立新的 Pareto frontier,表現勝過專用模型及大 2 至 12 倍的模型。模型亦支援選擇性載入模態編碼器、調整 vision token 預算、截短嵌入維度及量化權重,以配合資源受限環境。
Published on Oct 27
Authors:
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
Abstract
EmbeddingGemma 2 is an open multimodal embedding model that unifies text, images, audio, and video representations with efficient resource-flexible design and state-of-the-art benchmark performance.
We present EmbeddingGemma 2, an open multimodal embedding model which natively represents text, images, audio, video, and combinations thereof in a unified space. Our approach introduces improvements such as simplified per-layer embeddings and a multi-stage curriculum with modality and objective progression which ensure strong quality across unimodal, cross-modal, and mixed-modal evaluations. Notably, in comprehensive embedding benchmarks across all modalities, EmbeddingGemma 2 establishes a new Pareto frontier, outperforming both specialist models and models two to twelve times larger. Furthermore, our design offers flexibility for resource-constrained environments, supporting selective modality encoder loading, adjustable vision token budgets, truncatable embedding dimensions, and quantized weights. We present ablations analyzing key decisions and release EmbeddingGemma 2 to promote further research.
View arXiv page View PDF Add to collection
Get this paper in your agent:
hf papers read 2610.37364
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 1
Feature Extraction • 0.7B • Updated about 8 hours ago • 283k • 2.99kgoogle/embeddinggemma-2
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.37364 in a dataset README.md to link it from this page.
Spaces citing this paper 46
Browse 46 spaces citing this paper
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co