Showing cs.CLShow all
3 papers · 1 filter
cs.CL2026
Geometric Metrics and LLMs: What They Measure and When They Work
Viacheslav Yusupov, Anna Antipina, Ameliia Alaeva +6
We present a systematic stress-test of geometric metrics for LLM evaluation. Rank-based geometric properties of internal representations have shown promise as reference-free qualit…
cs.CL2025
Beyond Early-Token Bias: Model-Specific and Language-Specific Position Effects in Multilingual LLMs
Mikhail Menschikov, Alexander Kharitonov, Maiia Kotyga +5
Large Language Models (LLMs) exhibit position bias systematically underweighting information based on its location in the context but how this bias varies across languages and mode…
cs.CL2025
Token Homogenization under Positional Bias
Viacheslav Yusupov, Danil Maksimov, Ameliia Alaeva +9
This paper investigates token homogenization - the convergence of token representations toward uniformity across transformer layers and its relationship to positional bias in large…