Empirical analysis of a constrained language-model contest shows a 13.6% BPB improvement from 1.2244 to 1.058 through many minor optimizations, with most technique gains shrinking in top submissions.
Proceedings of the BabyLM Challenge at the Conference on Computational Natural Language Learning , year =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Parameter Golf: What Really Works?
Empirical analysis of a constrained language-model contest shows a 13.6% BPB improvement from 1.2244 to 1.058 through many minor optimizations, with most technique gains shrinking in top submissions.