VISTA 350M Collection VISTA 350M models from "No Single Tokenizer Feature Reliably Predicts Downstream Language Model Performance" (EMNLP 2026, Findings) • 33 items • Updated 28 days ago • 2
VISTA 130M Collection VISTA 130M models from "No Single Tokenizer Feature Reliably Predicts Downstream Language Model Performance" (EMNLP 2026, Findings) • 84 items • Updated 28 days ago • 4
Learning POMDP World Models from Observations with Language-Model Priors Paper • 2605.13740 • Published May 13 • 6
Global PIQA: Evaluating Physical Commonsense Reasoning Across 100+ Languages and Cultures Paper • 2510.24081 • Published Oct 28, 2025 • 24
Learning POMDP World Models from Observations with Language-Model Priors Paper • 2605.13740 • Published May 13 • 6