Paper Detail

Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM

Adam Zachary Wasserman, David Beauchemin

arxiv Score 7.8

Published 2026-09-15 · First seen 2026-09-17

General AI

Abstract

We submit MéTRON-FR, a 125M GPT-2 pretrained on 92.47M words of French, to the BabyLM 2026 Strict track. It scores 85.97 +/- 0.17% on QFrBLiMP (a native Quebec-French benchmark of grammatical minimal pairs) and 62.80% on the BabyLM-weighted leaderboard. A cross-lingual GLUE (General Language Understanding Evaluation) protocol that combines French task-data translation with rank-16 LoRA (Low-Rank Adaptation) produces a sharp task-type gradient: relational tasks gain measurably, while world-knowledge tasks regress. Bilingual Lexicon Induction aligns the French embeddings to GPT-2 at p@1 = 68.84 +/- 8.61%, 18X above chance, suggesting cross-lingual alignment tracks acquired grammatical competence rather than training duration. An ablation study shows that single-token zero-shot scoring is dominated by tokenizer and template artifacts at the child scale, motivating tokenizer-swap sensitivity, placebo-controlled prompting, and native-language minimal-pair benchmarks as standard diagnostics.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{wasserman2026right,
  title = {Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM},
  author = {Adam Zachary Wasserman and David Beauchemin},
  year = {2026},
  abstract = {We submit MéTRON-FR, a 125M GPT-2 pretrained on 92.47M words of French, to the BabyLM 2026 Strict track. It scores 85.97 +/- 0.17\% on QFrBLiMP (a native Quebec-French benchmark of grammatical minimal pairs) and 62.80\% on the BabyLM-weighted leaderboard. A cross-lingual GLUE (General Language Understanding Evaluation) protocol that combines French task-data translation with rank-16 LoRA (Low-Rank Adaptation) produces a sharp task-type gradient: relational tasks gain measurably, while world-knowle},
  url = {https://arxiv.org/abs/2609.17435},
  keywords = {cs.CL},
  eprint = {2609.17435},
  archiveprefix = {arXiv},
}

Metadata

{}