arXiv Computation and Language By Oskar Holmstr\"om, Marcel Bollmann, Marco Kuhlmann

A Systematic Comparison of Multilingual Interpretability Methods Reveals Anisotropy-Driven Failures

Read the original on arXiv Computation and Language →

The paper evaluates four metrics—CKA, ANC, GMM dominance per token, and ILO—used to measure cross‑lingual representation sharing in multilingual language models. Across 21 models ranging from 125 M to 14 B parameters, the metrics disagree, and the authors attribute this to anisotropy, where representations cluster in a narrow embedding cone. Only ILO shows a strong, robust correlation with cross‑lingual transfer performance (Spearman’s ρ = 0.90) after controlling for model size, family, and task variation, leading the authors to recommend ILO as the primary metric alongside anisotropy diagnostics.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv Computation and Language
Aug 28

Double Trouble: Bilingual Pretraining Leaves Language-Conditioned Effects in Shared-Language Representations

The study compares an English-only and a bilingual decoder-only model, each 310 M parameters, trained on eight diverse languages while controlling for English exposure, compute, and document overlap. After aligning on shared English vocabulary, the authors find that token embeddings appear similar, but the deeper hidden states used for prediction diverge across models. This hidden‑state mismatch grows through middle transformer layers and persists despite controls, indicating that contextual processing differs between the models. "whyItMatters":"The findings show that embedding alignment can conceal significant internal representation differences, which is crucial for any downstream work that assumes aligned multilingual models are interchangeable."

By Anjishnu Mukherjee, Ziwei Zhu, Antonios Anastasopoulos
arXiv AI
Sep 2

Lingua Franca or Probing Artifact? Rethinking Latent Language in Multilingual LLMs

The paper investigates whether different latent language probes—GMM-based representation probes and decoding-based probes—measure the same phenomenon in multilingual language models. Across various model families, training regimes, domains, tasks, checkpoints, and up to 27 languages, the authors find systematic disagreement: representation probes indicate earlier cross‑lingual mixing, while decoding probes reveal sharper, English‑biased language signals. These differences correlate with model multilinguality and training progression but remain relatively stable across domains, suggesting that current probes capture distinct aspects of multilingual processing rather than a single internal lingua franca.

By Deniz Bayazit, Badr AlKhamissi, Antoine Bosselut
arXiv Computation and Language
Aug 27

Apples to Apples? Towards Comparable Crosslingual Language Model Evaluation

The paper investigates how to fairly compare language models across languages, noting that current evaluation methods vary widely and lack empirical validation. By training controlled monolingual models on parallel data and testing multilingual LLMs, the authors find that many normalized metrics suffer from biases due to tokenization, encoding, and orthographic differences. Instead, they recommend using sentence‑level negative log‑likelihood over semantically equivalent sequences for more reliable cross‑lingual comparisons.

By Xiulin Yang, Ethan Gotlieb Wilcox, Catherine Arnett