AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

ML-Embed: Inclusive and Efficient Embeddings for a Multilingual World

Ziyin Zhang, Zihan Liao, Hang Yu, Peng Di, Rui Wang

ICML 2026 · Architectures and efficiency › RNNs and normalization · approach: Neural scaling laws · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add 3D-ML training before pruning to 120M (ablation)EuroBERT-210M · MTEB average (17 benchmarks)mean score: 44.10 -> 56.77Improved
add 3D-ML training before pruning to 120M (ablation)EuroBERT-210M · MTEB Multilingualmean score: 46.08 -> 56.94Improved
add MEL (nested-rank training) over baseline when compressing embedding at inference (ablation)Qwen3-0.6B embedding model · MTEB Englishmean score: 53.25 -> 64.30Improved
train with factorized embedding (decomposition-only, rank 512) versus original embedding (ablation)Qwen3-0.6B embedding model · MTEB Englishmean score: 69.68 -> 69.60No clear effect
add MEL on top of MLL (4 layers vs 1-layer baseline at same ~170M parameters) (ablation)Qwen3-0.6B embedding model · MTEB Englishmean score: +15 pointsImproved
add MLL (single multi-exit model) versus individually trained shallow models (ablation)Qwen3-0.6B embedding model · MTEB Englishmean scoreHurt
use ML-Embed curated data instead of KaLM-Embedding data (ablation)ML-Embed-0.6B · 17 MTEB benchmarksbenchmarks won: 9 out of 17 benchmarksImproved
use ML-Embed curated data instead of KaLM-Embedding data (ablation)ML-Embed-0.6B · MTEB Chinesemean scoreHurt
scale model from 140M to 8BML-Embed · MTEB averagemean score: 50.76 -> 70.24Improved
compare ML-Embed-8B with previous top-1 on PolishML-Embed-8B · MTEB Polishmean score: 50.95 -> 73.84Improved
compare ML-Embed-8B with previous top-1 on Multilingual/EnglishML-Embed-8B · MTEB Multilingualmean score: 72.32 -> 66.79Hurt
compare ML-Embed-0.6B with Qwen3-Embedding-0.6BML-Embed-0.6B · MTEB Multilingualmean score: 64.34 -> 61.30Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy