PulseAugur
EN
LIVE 08:41:03

LLM math reasoning shows partial cross-lingual parameter overlap

By PulseAugur Editorial · [2 sources] ·

A new research paper investigates the cross-lingual performance of large language models (LLMs) in mathematical reasoning. The study found that while there is some overlap in the parameters supporting mathematical abilities across languages, these parameters are not entirely shared. English-language models demonstrated the largest set of math-relevant parameters, with lower-resource languages showing smaller, though still overlapping, sets. AI

IMPACT This research could inform the development of more equitable multilingual LLMs by highlighting how mathematical reasoning is represented across different languages.

RANK_REASON The cluster contains an academic paper detailing research findings on LLM capabilities.

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.CL TIER_1 English(EN) · Behzad Shomali, Luisa Victor, Tim Selbach, Ali Hamza Bashir, David Berghaus, Joachim Koehler, Mehdi Ali, Markus Frey ·

    LLM Parameters for Math Across Languages: Shared or Separate?

    arXiv:2606.18453v1 Announce Type: new Abstract: Large language models (LLMs) exhibit substantial cross-lingual variation in mathematical reasoning performance, but it remains unclear whether these differences reflect language-specific parameters or a shared mechanism that manifes…

  2. arXiv cs.CL TIER_1 English(EN) · Markus Frey ·

    LLM Parameters for Math Across Languages: Shared or Separate?

    Large language models (LLMs) exhibit substantial cross-lingual variation in mathematical reasoning performance, but it remains unclear whether these differences reflect language-specific parameters or a shared mechanism that manifests differently by language. We present a cross-l…

RELATED ENTITIES

RELATED TOPICS