LANGSAE EDITING: Improving Multilingual Information Retrieval via Post-hoc Language Identity Removal

• URL: http://arxiv.org/abs/2601.04768v1
• Date: Thu, 08 Jan 2026 09:36:41 GMT
• Title: LANGSAE EDITING: Improving Multilingual Information Retrieval via Post-hoc Language Identity Removal
• Authors: Dongjun Kim, Jeongho Yoon, Chanjun Park, Heuiseok Lim,
• Abstract summary: multilingual embeddings encode language identity alongside semantics.<n>We propose LangSAE EDITING, a post-hoc sparse autoencoder trained on pooled embeddings.<n> Experiments across multiple languages show consistent improvements in ranking quality and cross-language coverage.
• Score: 34.73949500194166
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Dense retrieval in multilingual settings often searches over mixed-language collections, yet multilingual embeddings encode language identity alongside semantics. This language signal can inflate similarity for same-language pairs and crowd out relevant evidence written in other languages. We propose LANGSAE EDITING, a post-hoc sparse autoencoder trained on pooled embeddings that enables controllable removal of language-identity signal directly in vector space. The method identifies language-associated latent units using cross-language activation statistics, suppresses these units at inference time, and reconstructs embeddings in the original dimensionality, making it compatible with existing vector databases without retraining the base encoder or re-encoding raw text. Experiments across multiple languages show consistent improvements in ranking quality and cross-language coverage, with especially strong gains for script-distinct languages.

Related papers

• Tracing Multilingual Representations in LLMs with Cross-Layer Transcoders [51.380449540006985]
  Large Language Models (LLMs) can process many languages, yet how they internally represent this diversity remains unclear.<n>Do they form shared multilingual representations with language-specific decoding, and if so, why does performance still favor the dominant training language?<n>We analyze their internal mechanisms using cross-layer transcoders (CLT) and attribution graphs.
  arXiv  Detail & Related papers  (2025-11-13T22:51:06Z)
• Language Drift in Multilingual Retrieval-Augmented Generation: Characterization and Decoding-Time Mitigation [11.110312833458421]
  We study output language drift in multilingual RAG across multiple datasets, languages, and LLM backbones.<n>Our experiments reveal that the drift results from decoder-level collapse, where dominant token distributions and high-frequency English patterns dominate the intended generation language.<n>We propose Soft Constrained Decoding (SCD), a lightweight, training-free decoding strategy that gently steers generation toward the target language.
  arXiv  Detail & Related papers  (2025-11-13T05:36:31Z)
• DIVERS-Bench: Evaluating Language Identification Across Domain Shifts and Code-Switching [8.14614722074297]
  Language Identification (LID) is a core task in multilingual NLP, yet current systems often overfit to clean, monolingual data.<n>This work introduces DIVERS-BENCH, a comprehensive evaluation of state-of-the-art LID models across diverse domains.<n>Our findings reveal that while models achieve high accuracy on curated datasets, performance degrades sharply on noisy and informal inputs.
  arXiv  Detail & Related papers  (2025-09-22T13:32:31Z)
• Discovering Low-rank Subspaces for Language-agnostic Multilingual Representations [38.56175462620892]
  Large pretrained multilingual language models (ML-LMs) have shown remarkable capabilities of zero-shot cross-lingual transfer. We present a novel view of projecting away language-specific factors from a multilingual embedding space. We show that applying our method consistently leads to improvements over commonly used ML-LMs.
  arXiv  Detail & Related papers  (2024-01-11T09:54:11Z)
• AdaCCD: Adaptive Semantic Contrasts Discovery Based Cross Lingual Adaptation for Code Clone Detection [69.79627042058048]
  AdaCCD is a novel cross-lingual adaptation method that can detect cloned codes in a new language without annotations in that language. We evaluate the cross-lingual adaptation results of AdaCCD by constructing a multilingual code clone detection benchmark consisting of 5 programming languages.
  arXiv  Detail & Related papers  (2023-11-13T12:20:48Z)
• The Geometry of Multilingual Language Model Representations [25.880639246639323]
  We assess how multilingual language models maintain a shared multilingual representation space while still encoding language-sensitive information in each language. The subspace means differ along language-sensitive axes that are relatively stable throughout middle layers, and these axes encode information such as token vocabularies. We visualize representations projected onto language-sensitive and language-neutral axes, identifying language family and part-of-speech clusters, along with spirals, toruses, and curves representing token position information.
  arXiv  Detail & Related papers  (2022-05-22T23:58:24Z)
• FILTER: An Enhanced Fusion Method for Cross-lingual Language Understanding [85.29270319872597]
  We propose an enhanced fusion method that takes cross-lingual data as input for XLM finetuning. During inference, the model makes predictions based on the text input in the target language and its translation in the source language. To tackle this issue, we propose an additional KL-divergence self-teaching loss for model training, based on auto-generated soft pseudo-labels for translated text in the target language.
  arXiv  Detail & Related papers  (2020-09-10T22:42:15Z)
• Inducing Language-Agnostic Multilingual Representations [61.97381112847459]
  Cross-lingual representations have the potential to make NLP techniques available to the vast majority of languages in the world. We examine three approaches for this: (i) re-aligning the vector spaces of target languages to a pivot source language; (ii) removing language-specific means and variances, which yields better discriminativeness of embeddings as a by-product; and (iii) increasing input similarity across languages by removing morphological contractions and sentence reordering.
  arXiv  Detail & Related papers  (2020-08-20T17:58:56Z)
• XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning [68.57658225995966]
  Cross-lingual Choice of Plausible Alternatives (XCOPA) is a typologically diverse multilingual dataset for causal commonsense reasoning in 11 languages. We evaluate a range of state-of-the-art models on this novel dataset, revealing that the performance of current methods falls short compared to translation-based transfer.
  arXiv  Detail & Related papers  (2020-05-01T12:22:33Z)
• Bridging Linguistic Typology and Multilingual Machine Translation with Multi-View Language Representations [83.27475281544868]
  We use singular vector canonical correlation analysis to study what kind of information is induced from each source. We observe that our representations embed typology and strengthen correlations with language relationships. We then take advantage of our multi-view language vector space for multilingual machine translation, where we achieve competitive overall translation accuracy.
  arXiv  Detail & Related papers  (2020-04-30T16:25:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.