ldwang

ldwang on Hugging Face Daily Papers: 9 papers, 2 in the top 3 of their day, 318 upvotes.

1. CoWindow Attention: Full Causal Coverage Is a Collective Property 67 upvotes, #10 of 2026-09-29
2. MassAlloc Attention: Let Attention Allocate Its Own Compute 75 upvotes, #8 of 2026-09-29
3. Trainable Dynamic Mask Sparse Attention 14 upvotes, #5 of 2025-08-04
4. Infinity Instruct: Scaling Instruction Selection and Synthesis to Enhance Language Models 5 upvotes, #21 of 2025-06-16
5. CCI4.0: A Bilingual Pretraining Dataset for Enhancing Reasoning in Large Language Models 9 upvotes, #20 of 2025-06-10
6. Infinity-MM: Scaling Multimodal Performance with Large-Scale and High-Quality Instruction Data 17 upvotes, #6 of 2024-10-28
7. CCI3.0-HQ: a large-scale Chinese dataset of high quality designed for pre-training large language models 8 upvotes, #17 of 2024-10-25
8. Emu3: Next-Token Prediction is All You Need 75 upvotes, #1 of 2024-09-30
9. Aquila2 Technical Report 12 upvotes, #2 of 2024-08-15

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.