Cross-Lingual Transfer

Follow this topic Ask about this topic

Latest papers 239

30 papers in the last four weeks, up 25% on the four weeks before. 0.2% of all new papers.

All topics

Cross-Lingual Transfer of Evil Vectors Is Directionally Asymmetric
Oct 13, 2026Juann Kim, Minku KangCross-Lingual Transfer Activation Steering
University of Southern California · KAIST
Does System One Transfer? Calibrated Typed-Decision Models, Label Budgets, and Translation for Indonesian Regional Languages
Oct 9, 2026Romi Fadli, Rusdah, Wahyu Ramadani Putri Utami +1Multilingual Language Model Evaluation Cross-Lingual Transfer
Faculty of Information Technology, Universitas Budi Luhur, Jakarta, Indonesia
Adapting English Quality Classifiers for Multilingual LLM Pretraining Data Selection
Oct 8, 2026Vinko Sabolčec, Bettina Messmer, Yassine Turki +1Language Model Pretraining Cross-Lingual Transfer
EPFL
Cross-lingual Calibration of Pre-Generation Success Probes for Multilingual LLM Routing
Oct 5, 2026Andrea Paganelli, Stefano Civelli, Pietro Bernardelle +1Confidence Estimation in Language Models Language Model Calibration
The University of Queensland · Polytechnic University of Milan
Cross-Lingual Transferability of Training Data Extraction Attacks to Recover Memorized PII
Oct 5, 2026Alexandru Nazare, Agnese Profico, Nicolò Vania +6Multilingual Language Models Cross-Lingual Representation Alignment
Human-Centric ART, University of Rome Tor Vergata · Department of Computer Science, University of Luxembourg · Almawave Labs, Rome
Multilingual GSM-Symbolic: What determines capability transfer across languages?
Oct 2, 2026Kenneth Enevoldsen, Riley Herchert, Sofie Mosegaard +22Mathematical Reasoning Benchmarks Multilingual Language Model Evaluation
Aarhus University · Danish Foundation Models · University of Alabama +13
The Score Is Not the Structure: Brain Alignment and Cross-Lingual Transfer
Oct 2, 2026Saman RahbarCross-Lingual Transfer Cross-Modal Alignment
Independent researcher Vancouver, Canada
Query-aware routing for Cross-lingual performance gains in Encoders
Oct 2, 2026Akshay Jain, Edward KimMultilingual IR Cross-Lingual Transfer
ConfidentialMind
Cross-Lingual Alignment for Decoder-Only Models using MoE Routers
Oct 1, 2026Lucas Bandarkar, Clark Peng, Ahmed Haj Ahmed +2Mixture-of-Experts Language Models Cross-Lingual Representation Alignment
University of California, Los Angeles∗ · Haverford College · MILA - Quebec AI Institute & McGill University
MGhana-ST: A Low-Resource Speech Translation Dataset for Ghanaian Languages and an Analysis of Multilingual Training Trade-offs
Sep 30, 2026Frank Lawrence Nii Adoquaye Acquaye, Eric George Parakal, Jesse Johnson +2Speech Translation Low-Resource Language Processing
Ashesi University, Berekuso, Ghana · AdwumaTech AI, Accra, Ghana · HSE University, Moscow, Russia +1
Zero-Compute Cross-Lingual Transferability Estimation Using Typological Feature Proxies
Sep 30, 2026Dalton Raphael Harmsen, Swier Garst, Thomas van Osch +2Task Transferability Estimation Linguistic Typology
AMOR/e Lab, Eindhoven University of Technology, Eindhoven, The Netherlands · SURF, Amsterdam, The Netherlands
BARRAC: Adaptation of an English Aspect-based Sentiment Analysis Approach for Classification Tasks in Arabic Dialects
Sep 30, 2026Ali Almutairi, Gelareh Mohammadi, Imran Razzak +1Arabic NLP Sentiment Analysis
University of New South Wales, Australia · MBZUAI, UAE
Zero-shot Dependency Parsing with Unsupervised Cross-Lingual Bootstrapping
Sep 29, 2026Lalita Lowphansirikul, Attapol Rutherford, Jian Gang Ngui +2Multilingual Language Models Low-Resource Language Processing
VISTEC, Thailand · Chulalongkorn University, Thailand · AI Singapore, Singapore
How Many Labels Does a Language Need? Annotation Budgets and Cross-Lingual Pooling for African-Language Text Classification
Sep 29, 2026Bhanu Prakash Vangala, Sowmya Guda, Navya VangalaZero-Shot Text Classification Text Classification
University of Missouri · Amar Bio Tech Pvt Ltd
RunyaNER: Auxiliary Language Selection for Runyankore NER
Sep 29, 2026Prosper Arineitwe Asiimwe, Francois Meyer, Jan BuysNamed Entity Recognition Multilingual Language Model Evaluation
Department of Computer Science, University of Cape Town
Language Distances are Practical for Equitable Cross-Lingual Transfer
Sep 26, 2026York Hay Ng, Razan Ahsan Rifandi, Aditya Khan +1Cross-Lingual Transfer Low-Resource Language Processing
University of Toronto, Canada · Ontario Tech University, Canada
Cross-Scale Transfer Learning for Depression Severity Prediction: From PHQ-8 to HAMD-17 Across Languages and Clinical Paradigms
Sep 23, 2026Wenjie Feng, Sahba Zojaji, Satoshi NakamuraClinical Outcome Prediction Depression Severity Estimation
School of Artificial Intelligence, The Chinese University of Hong Kong, Shenzhen, Shenzhen, China · School of Humanities and Social Science, The Chinese University of Hong Kong, Shenzhen, Shenzhen, China · Shenzhen Loop Area Institute, Shenzhen, China
Isolated Sign Language Recognition for Icelandic Sign Language: Experiments in a Low-resource Setting
Sep 22, 2026Finnur Ágúst Ingimundarson, Guðný Björk Þorvaldsdóttir, Mathias Müller +1Few-Shot Learning Sign Language Recognition
Department of Computational Linguistics, University of Zurich · Communication Centre for the Deaf and Hard of Hearing in Iceland
Rethinking Pivot Programming Languages in Code Language Models
Sep 19, 2026Andor Diera, Lukas Galke Poech, Matthias TichyCode Language Models Cross-Lingual Transfer
Ulm University Ulm, Germany · University of Southern Denmark Odense, Denmark
Beetle: A Bilingual Model Suite for Modelling Second-Language Processing
Sep 18, 2026Suchir Salhan, Catherine Arnett, James Michaelov +1Multilingual Language Models Language Model Pretraining
University of Cambridge · EleutherAI · MIT
Improving Cross-Lingual Transfer for Sequential Sentence Classification in Research Papers via Structural Similarity
Sep 17, 2026Kazuhiro Yamauchi, Marie KatsuraiDocument Understanding Text Classification
Doshisha University Kyotanabe, Japan
A Cross-Lingual Acoustic Disease-Alignment Framework for Respiratory Health Assessment from Spontaneous Speech
Sep 16, 2026Roksana Khanom, Raghib Asfak Tasnim, Bodrun Nahar Bithi +4Cross-Lingual Representation Alignment Speech Processing
University of Maryland, College Park, USA · Sylhet MAG Osmani Medical College Hospital, Bangladesh · Dr. M R Khan Shishu Hospital & Institute of Child Health, Bangladesh +1
Why Pretraining Fails to Share Cross-Lingual Knowledge
Sep 16, 2026Adam Gaber, Uriel Dolev, Elisabeth Fittschen +3Language Model Pretraining Multilingual Language Models
Weizmann Institute of Science · Bar-Ilan University · Johns Hopkins University +4
Zero-Shot Cross-Lingual Recognition of Sign Language Handshapes
Sep 16, 2026Marcel Granero-Moya, Carolina del Corral Farrarós, Gloria Haro +2Zero-Shot Learning Sign Language Recognition
Universitat Pompeu Fabra
Cross-Lingual Parkinson's Disease Severity Assessment Using Pre-trained Speech Embeddings: A Multi-Class Evaluation
Sep 16, 2026Simon Pals, Cristian Tejedor-GarciaParkinson's Disease Clinical Speech Processing
Centre for Language Studies Radboud University Nijmegen, the Netherlands
Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM
Sep 15, 2026Adam Zachary Wasserman, David BeaucheminLLM Evaluation Cross-Lingual Representation Alignment
1Open Honest Foundation · 2Group for Research in Artificial Intelligence of Laval University (GRAIL), Université Laval, Québec, Canada
Language Orthogonalization for Zero-Shot Cross-Lingual Audio Deepfake Detection
Sep 15, 2026Minu Kim, Ji Sub Um, Hoirin KimAudio Deepfake Detection Invariant Representation Learning
University of Southern California, USA · KAIST, South Korea
Efficient One-to-Many Translation with Joint Multi-Stream Diffusion
Sep 14, 2026Yiwen Guan, Jacob WhitehillDiffusion Language Model Inference Diffusion Language Models
Worcester Polytechnic Institute
Sequential Adapter Stacking for Cross-Lingual Low-Resource ASR
Sep 14, 2026Thai Thi Thanh Thao Dang, Mengjie Qian, Kate KnillLow-Resource Speech Recognition Parameter-Efficient Fine-Tuning
Department of Engineering, University of Cambridge, UK
Cross-Lingual F5-TTS 2: A Simplified Framework for Language-Agnostic Voice Cloning
Sep 14, 2026Qingyu Liu, Rixi Xu, Yushen Chen +11TTS Synthesis Zero-Shot TTS
Center for Language and Speech Processing, Johns Hopkins University, Baltimore, MD, USA · MoE Key Lab of Artificial Intelligence, X-LANCE Lab, Shanghai Jiao Tong University, Shanghai, China · Shanghai Innovation Institute, Shanghai, China +3

Oct 13, 2026J/K move · Enter open · S save

Cross-Lingual Transfer of Evil Vectors Is Directionally Asymmetric

Juann Kim, Minku Kang

University of Southern California · KAIST

Cross-Lingual Transfer Activation Steering Cross-Lingual Representation Alignment Multilingual Language Models

Open paper AskPDF arXiv

Persona vectors are linear directions in the residual-stream activations of LLMs that encode behavioral traits and enable fine-grained control over model behavior. Among them, the evil persona vector is particularly relevant to understanding harmful behavior, jailbreak vulnerabilities, and the unlearning of harmful knowledge. Yet whether this vector reflects a genuinely cross-lingual concept, rather than an artifact of its extraction language, remains untested. We conduct a systematic evil-vector transfer study across two high-resource languages (English, Spanish) and two low-resource languages (Swahili, Telugu) on four open-source instruction-tuned LLMs, and find that transfer is directionally asymmetric. Evil vectors are harder to obtain reliably from low-resource languages, and the resulting vectors are largely ineffective at steering, even within their own language. In contrast, vectors extracted from high-resource languages transfer effectively to low-resource inputs while largely preserving general knowledge, as measured by MMLU. Geometric analysis further supports this pattern: evil vectors from the two high-resource languages are closely aligned, whereas low-resource vectors align only weakly with this shared direction. These results suggest that a shared cross-lingual evil direction exists and, once reliably extracted from a high-resource language, can be reused to steer behavior in low-resource languages without additional per-language data collection.