CoolFace
Modelpublic

amosluna/babylm-2026-pstar-seed42

sourceHugging Faceapache-2.0updated 4mo agoView on Hugging Face
0likes6downloads
Model Card

TAAM — Typology-Aware Adaptive Mixing — Pstar seed 42

This is a BabyLM 2026 Multilingual Track submission checkpoint. It was trained on English + Dutch + Mandarin Chinese under a ≤100M-unique-token budget with the TAAM method.

  • —Method: Pstar
  • —Seed: 42
  • —Repo: amosluna/babylm-2026-pstar-seed42
  • —Final π (per-language sampling probability): eng=0.159, nld=0.258, zho=0.583
  • —Total token exposures: 655360000
  • —Training wall-clock: 10126.297962903976 s
  • —Source run dir: runs/2026-06-01_Pstar_seed42

Intermediate checkpoints

This repo exposes 24 intermediate checkpoints as branches following the BabyLM 2026 naming convention: chck_1M, chck_2M, ..., chck_10M, chck_20M, ..., chck_100M, chck_200M, ..., chck_600M. The eval pipeline at babylm-org/babylm-eval pulls these revisions automatically with:

bash
bash multilingual/scripts/zeroshot_model_fast_all.sh \
    --model_name amosluna/babylm-2026-pstar-seed42

Usage

python
from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("amosluna/babylm-2026-pstar-seed42")
model = AutoModelForCausalLM.from_pretrained("amosluna/babylm-2026-pstar-seed42")  # final checkpoint
# Intermediate checkpoint:
# model = AutoModelForCausalLM.from_pretrained("amosluna/babylm-2026-pstar-seed42", revision="chck_100M")

Method summary

TAAM combines (a) a URIEL/lang2vec-derived typological prior over initial sampling probabilities, (b) EXP3 online updates over per-language sampling probabilities, and (c) byte-premium-aware token budgeting. The two reward variants are normalized_excess_loss (v1, delta-based) and cross_lingual_deficit (v2, level-based).

See the paper and the public repo for full details, including the structural floor derivation that explains why the v1 reward starves the hardest language under typological asymmetry.

Citation

bibtex
@inproceedings{taam2026,
  title  = {Typology-Aware Adaptive Mixing for Multilingual BabyLMs},
  author = {Luna, Amos and collaborators},
  year   = {2026},
  booktitle = {Proceedings of the BabyLM Workshop at EMNLP 2026}
}