#Iclr2023

keyu-tian

SparK

[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"

yinizhilian

ICLR2025-Papers-with-Code

历年ICLR论文和开源项目合集，包含ICLR2021、ICLR2022、ICLR2023、ICLR2024、ICLR2025.

megvii-research

RevCol

Official Code of Paper "Reversible Column Networks" "RevColv2"

jrobine

twm

Transformer-based World Models

matt-seb-ho

WikiWhy

WikiWhy is a new benchmark for evaluating LLMs' ability to explain between cause-effect relationships. It is a QA dataset containing 9000+ "why" question-answer-rationale triplets.