Long Document Summarization Papers
The official code of EMNLP 2022, "How Far are We from Robust Long Abstractive Summarization?".
[EMNLP'26] Agentic Visual Perception for Long-Document Understanding
PaperDoctor 是一个用于科研论文审阅与修改的 AI Agent。
[EMNLP 2025] M-LongDoc: A Benchmark For Multimodal Super-Long Document Understanding And A Retrieval-Aware Tuning Framework
A Very long (>10,000 words) version of scientific papers dataset
compiles and parses the summarization gauntlet and results from various models into a dataset-like format
Style change detection dataset using AO3 fics
AI-powered long-document editor. Clean Word ↔ Markdown workflows. Local-first, private, and built for professionals.
Classify documents longer than your transformer's context window: overlapping windows, cross-window aggregation (max / mean / top-k / log-mean-exp), and window-level gradient checkpointing that cuts training memory. Encoder-agnostic and task-agnostic, with measured speed and accuracy trade-offs.
To associate your repository with the long-document topic, visit your repo's landing page and select "manage topics."