WikiText-2

Wikitext-2 is a dataset used to evaluate the performance of language models, containing a collection of text from Wikipedia articles.

Papers using WikiText-2 (4)

KronQ: LLM Quantization via Kronecker-Factored Hessian2026LACE-SVD: Loss-Aware SVD with Cumulative Error Correction for LLM Compression2026SA-ADP: Sensitivity-Aware Adaptive Differential Privacy for Large Language Models2025SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights2025