Cover Image

Beyond Early-Token Bias: Model-Specific and Language-Specific Position Effects in Multilingual LLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Menschikov, Mikhail, Kharitonov, Alexander, Kotyga, Maiia, Porvatov, Vadim, Zhukovskaya, Anna, Kagramanyan, David, Shvetsov, Egor, Burnaev, Evgeny
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

How to model Human Actions distribution with Event Sequence Data
by: Surkov, Egor, et al.
Published: (2025)

Investigating the Impact of Quantization Methods on the Safety and Reliability of Large Language Models
by: Kharinaev, Artyom, et al.
Published: (2025)

GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs
by: Zhelnin, Maxim, et al.
Published: (2024)

PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents
by: Menschikov, Mikhail, et al.
Published: (2026)

EBES: Easy Benchmarking for Event Sequences
by: Osin, Dmitry, et al.
Published: (2024)

Token Homogenization under Positional Bias
by: Yusupov, Viacheslav, et al.
Published: (2025)

QuantNAS for super resolution: searching for efficient quantization-friendly architectures against quantization noise
by: Shvetsov, Egor, et al.
Published: (2022)

MLEM: Generative and Contrastive Learning as Distinct Modalities for Event Sequences
by: Moskvoretskii, Viktor, et al.
Published: (2024)

PersonalAI: A Systematic Comparison of Knowledge Graph Storage and Retrieval Approaches for Personalized LLM agents
by: Menschikov, Mikhail, et al.
Published: (2025)

From Internal Representations to Text Quality: A Geometric Approach to LLM Evaluation
by: Yusupov, Viacheslav, et al.
Published: (2025)

SeqNAS: Neural Architecture Search for Event Sequence Classification
by: Udovichenko, Igor, et al.
Published: (2024)

From 2:4 to 8:16 sparsity patterns in LLMs for Outliers and Weights with Variance Correction
by: Maximov, Egor, et al.
Published: (2025)

How does a Language-Specific Tokenizer affect LLMs?
by: Seo, Jean, et al.
Published: (2025)

Suing online platforms for copyright infringements: the choice of court and law in the “Project Gutenberg” scenario
by: Maiia Otchenash
Published: (2019)

Motivating Next-Gen Accelerators with Flexible (N:M) Activation Sparsity via Benchmarking Lightweight Post-Training Sparsification Approaches
by: Alanova, Shirin, et al.
Published: (2025)

Beyond Literal Token Overlap: Token Alignability for Multilinguality
by: Hämmerl, Katharina, et al.
Published: (2025)

Neural network atmospheric bias correction on heterogeneous data with fine‐scale dynamics preservation
by: Viktor Golikov, et al.
Published: (2026)

Light Schrödinger Bridge
by: Korotin, Alexander, et al.
Published: (2023)

The Density of Cross-Persistence Diagrams and Its Applications
by: Mironenko, Alexander, et al.
Published: (2026)

Qtok: A Comprehensive Framework for Evaluating Multilingual Tokenizer Quality in Large Language Models
by: Chelombitko, Iaroslav, et al.
Published: (2024)

Revising deep learning methods in parking lot occupancy detection
by: Martynova, Anastasia, et al.
Published: (2023)

T-3DGS: Removing Transient Objects for 3D Scene Reconstruction
by: Markin, Alexander, et al.
Published: (2024)

The Token Tax: Systematic Bias in Multilingual Tokenization
by: Lundin, Jessica M., et al.
Published: (2025)

Learning of Population Dynamics: Inverse Optimization Meets JKO Scheme
by: Persiianov, Mikhail, et al.
Published: (2025)

Draft (work in progress) Metric Topological Relaxation: A Physical Mechanism for Cosmological Coupling k ≈ 3 in Black Holes
by: Kharitonov, Dmitry
Published: (2026)

Integration of Neural Network Technologies into the Methodology of Teaching Sensory Analysis for Food Technology Students
by: Fedash, Maiia V.
Published: (2026)

ELO: Efficient Layer-Specific Optimization for Continual Pretraining of Multilingual LLMs
by: Yoo, HanGyeol, et al.
Published: (2026)

Online Algorithm for Aggregating Experts' Predictions with Unbounded Quadratic Loss
by: Korotin, Alexander, et al.
Published: (2025)

Mixability of Integral Losses: a Key to Efficient Online Aggregation of Functional and Probabilistic Forecasts
by: Korotin, Alexander, et al.
Published: (2019)

Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs
by: Nadeem, Afrozah, et al.
Published: (2026)

Language Bias under Conflicting Information in Multilingual LLMs
by: Östling, Robert, et al.
Published: (2026)

Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models
by: Tang, Tianyi, et al.
Published: (2024)

KEY AREAS OF USE OF REMOTE SERVERS IN THE ACTIVITIES OF ECONOMIC ENTITIES
by: Zhukovskaya Irina Evgenyevna, Dadabayeva Rano Akramovna
Published: (2025)

Challenges of Multilingual Program Specification and Analysis
by: Furia, Carlo A., et al.
Published: (2024)

Light Unbalanced Optimal Transport
by: Gazdieva, Milena, et al.
Published: (2023)

Light and Optimal Schrödinger Bridge Matching
by: Gushchin, Nikita, et al.
Published: (2024)

Q-RAG: Long Context Multi-step Retrieval via Value-based Embedder Training
by: Sorokin, Artyom, et al.
Published: (2025)

How Language Directions Align with Token Geometry in Multilingual LLMs
by: Kim, JaeSeong, et al.
Published: (2025)

Framing Political Bias in Multilingual LLMs Across Pakistani Languages
by: Nadeem, Afrozah, et al.
Published: (2025)

Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment
by: Banerjee, Somnath, et al.
Published: (2025)