# Retrieval Disagreement as a Signal: Diagnosing and Steering Multilingual Dense Embedders

arXiv: https://arxiv.org/abs/2610.18526 · HTML: https://mlsanity.com/paper/2610.18526/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 37 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Builds on (most cited first, 27 of 27)

- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://mlsanity.com/paper/2212.03533/) (2022)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation](https://mlsanity.com/paper/2004.09813/) (2020)
- [On the Cross-lingual Transferability of Monolingual Representations](https://mlsanity.com/paper/1910.11856/) (2019)
- [Word Translation Without Parallel Data](https://mlsanity.com/paper/1710.04087/) (2017)
- [Multilingual E5 Text Embeddings: A Technical Report](https://mlsanity.com/paper/2402.05672/) (2024)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [A Survey Of Cross-lingual Word Embedding Models](https://mlsanity.com/paper/1706.04902/) (2017)
- [When and Why are Pre-trained Word Embeddings Useful for Neural Machine Translation?](https://mlsanity.com/paper/1804.06323/) (2018)
- [On the Language Neutrality of Pre-trained Multilingual Representations](https://mlsanity.com/paper/2004.05160/) (2020)
- [LAReQA: Language-agnostic answer retrieval from a multilingual pool](https://mlsanity.com/paper/2004.05484/) (2020)
- [Language-Specific Latent Process Hinders Cross-Lingual Performance](https://mlsanity.com/paper/2505.13141/) (2025)
- [Discovering Low-rank Subspaces for Language-agnostic Multilingual Representations](https://mlsanity.com/paper/2401.05792/) (2024)
- [A Simple and Effective Method To Eliminate the Self Language Bias in Multilingual Representations](https://mlsanity.com/paper/2109.04727/) (2021)
- [Steering into New Embedding Spaces: Analyzing Cross-Lingual Alignment Induced by Model Interventions in Multilingual Language Models](https://mlsanity.com/paper/2502.15639/) (2025)
- [Multi-EuP: The Multilingual European Parliament Dataset for Analysis of Bias in Information Retrieval](https://mlsanity.com/paper/2311.01870/) (2023)
- [The Cross-Lingual Cost: Retrieval Biases in RAG over Arabic-English Corpora](https://mlsanity.com/paper/2507.07543/) (2025)
- [Cross-Lingual Activation Steering for Multilingual Language Models](https://mlsanity.com/paper/2601.16390/) (2026)
- [Language Bias in Information Retrieval: The Nature of the Beast and Mitigation Methods](https://mlsanity.com/paper/2509.06195/) (2025)
- [All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG](https://mlsanity.com/paper/2604.20199/) (2026)
- [LANGSAE EDITING: Improving Multilingual Information Retrieval via Post-hoc Language Identity Removal](https://mlsanity.com/paper/2601.04768/) (2026)
- [Latent-Space Intervention for Cross-Lingual Factual Consistency: Consistency Improvements without Accuracy Drops](https://mlsanity.com/paper/2608.28860/) (2026)
- [Hubness, Not Anisotropy, Drives Cross-Lingual Retrieval Asymmetry in Multilingual Embedding Models](https://mlsanity.com/paper/2605.26575/) (2026)
- [SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval](https://mlsanity.com/paper/2606.18801/) (2026)

## Abstract (arXiv)

Multilingual dense retrievers place all languages in one shared space so that a question finds relevant documents, regardless of which language they are written in. We show that the same question, asked in different languages, retrieves substantially different documents, and this disagreement is not because of weak language understanding. Each language performs well on its own and ranks candidates consistently with the other languages; divergence appears only when languages share an index. We trace the mechanism: each language keeps its own ranking, but in the shared index, its correct answers are pushed far down, buried under non-relevant documents that are either in the query's own language or in English, depending on the model. We introduce disagreement-guided steering: for each language, we contrast the documents that other languages agree on for a question against those that language retrieves idiosyncratically, and take the difference of their representations as a language-specific direction, added to that language's queries at inference. Against baselines of language-mean removal, subspace removal, and per-language rotation, it most strongly reduces same-language crowding while remaining competitive on recall, trading off against mean removal, which better preserves ranking quality.