MIMO: Multilingual Information Retrieval via Monolingual Objectives
cs.IR, cs.AI
Submitted: 2026-05-29
Updated: 2026-08-27
Comments: EMNLP 2026 Main
Code: https://github.com/embeddings-benchmark/mteb
License: http://creativecommons.org/licenses/by/4.0/
The gist: Multilingual Information Retrieval (MLIR) reflects real-world search environments in which queries and relevant documents may appear in different languages within a mixed-language corpus.
Terminology
Abstract
Multilingual Information Retrieval (MLIR) reflects real-world search environments in which queries and relevant documents may appear in different languages within a mixed-language corpus. However, existing embedding models are primarily optimized for Multi-Monolingual retrieval and their performance often degrades in MLIR settings. Moreover, directly applying conventional contrastive learning to MLIR can exacerbate language clustering and expose a trade-off between cross-lingual alignment and embedding uniformity. To address these limitations, we propose MIMO: Multilingual Information Retrieval via Monolingual Objectives, a two-stage framework that uses a stable English semantic space from a high-performing teacher model as an anchor. MIMO first initializes the student model's cross-lingual alignment through knowledge distillation, and then jointly optimizes distillation and cross-lingual contrastive learning to improve retrieval discrimination while preserving alignment. Extensive experiments show that MIMO consistently outperforms existing cross-lingual training baselines across various MLIR and Multi-Monolingual benchmarks. MIMO also remains competitive with off-the-shelf models of similar or larger parameter scales. Furthermore, our cross-lingual Alignment-Uniformity analysis clarifies the distinct roles of the two loss components and shows that their combination yields a favorable trade-off between alignment and uniformity.
Sources
- On the Cross-lingual Transferability of Monolingual Representations
- Llama-Embed-Nemotron-8B: A Universal Text Embedding Model for Multilingual and Cross-Lingual Tasks
- mMARCO: A Multilingual Version of the MS MARCO Passage Ranking Dataset
- M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation
- Unsupervised Cross-lingual Representation Learning at Scale
- MMTEB: Massive Multilingual Text Embedding Benchmark
- jina-embeddings-v5-text: Task-Targeted Embedding Distillation
- Diffusion-Pretrained Dense and Contextual Embeddings
- Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup
- MTEB: Massive Text Embedding Benchmark
- Training Sparse Mixture Of Experts Text Embedding Models
- EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval
- Overview of the TREC 2022 NeuCLIR Track
- Overview of the TREC 2023 NeuCLIR Track
- NeuCLIRBench: A Modern Evaluation Collection for Monolingual, Cross-Language, and Multilingual Information Retrieval
- Towards General Text Embeddings with Multi-stage Contrastive Learning
- mmBERT: A Modern Multilingual Encoder with Annealed Language Learning
- Multilingual E5 Text Embeddings: A Technical Report
- Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
Related papers
- The Price of Isolation: Estimating the Ecosystem Cost of Symmetric Two-Sided A/B Testing
- SCAR: Semantic Continuity-Aware Retrieval for Efficient Context Expansion in RAG
- MixLoRA-DSI: Dynamically Expandable Mixture-of-LoRA Experts for Rehearsal-Free Generative Retrieval over Dynamic Corpora
- RRCM: Ranking-Driven Retrieval over Collaborative and Meta Memories for LLM Recommendation
- Right Family, Wrong Skill: Evaluating Risk Exposure in Agent Skill Retrieval
- UltRAG: a Universal Simple Scalable Recipe for Knowledge Graph RAG