Stars
Temporal Discrepancy Tomography (TDT) - AI-generated text detection via temporal tomography
Official style files for papers submitted to venues of the Association for Computational Linguistics
The Open-Source Data Annotation Platform
PyTorch implementation of "Supervised Contrastive Learning" (and SimCLR incidentally)
Code for "A Principled Framework for Multi-View Contrastive Learning"
Pytorch implementation of a RGCN Link Prediction Model
Source code for “Neural RST-based Evaluation of Discourse Coherence” (AACL-IJCNLP 2020)
Graph Convolutional Networks for Text Classification. AAAI 2019
Code repository for Kim et al's ACL 2024 paper: "Threads of Subtlety: Detecting Machine-Generated Texts Through Discourse Motifs"
A minimalist SOTA LaTeX OCR model with only 20M parameters, running in browser. Full training pipeline available for self-reproduction. | 超轻量SOTA LaTeX公式识别模型，仅20M参数量，可在浏览器中运行。训练全流程代码开源，以便自学复现。
An intuitive and low-overhead instrumentation tool for Python
Official implementation of "Continuous Autoregressive Language Models"
一款桌面应用程序，可将格式混乱的 Word 文档（及兼容格式）一键转换为符合规范的专业文档，实现排版工作自动化。
Official PyTorch implementation for "Large Language Diffusion Models"
A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
LUKE -- Language Understanding with Knowledge-based Embeddings
Large Concept Models: Language modeling in a sentence representation space
一个为Gemini对话添加交互式时间轴的浏览器扩展，让长对话导航和回顾更轻松。
A terminal-based SSH manager inspired by lazydocker and k9s - Written in go
A flexible framework powered by ComfyUI for generating personalized Nobel Prize images.
Asyncer, async and await, focused on developer experience.
Modern MD Editor 是一款面向创作者与内容团队的「高颜值 Markdown 编辑器 + 社交平台发布器」。它以极致的界面与交互为基础，提供所见即所得的实时预览与多端视口切换，并通过一键复制将 Markdown 转为适配微信公众号/社交平台的高保真富文本（自动内联样式、字体/行高/字距与主题化适配），让创作到发布的最后一步变得优雅、高效、可控。
Toolkit to segment text into sentences or other semantic units in a robust, efficient and adaptable way.