Yu Zhao

yuzhaouoe's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 2 months ago

Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression

Paper • 2503.02812 • Published Mar 4 • 9

upvoted a collection 2 months ago

Q-Filters

Collection

Pre-computed Q-Filters for efficient KV cache compression. • 15 items • Updated Mar 3 • 7

upvoted a paper 3 months ago

Competitive Programming with Large Reasoning Models

Paper • 2502.06807 • Published Feb 3 • 70

upvoted 4 papers 7 months ago

DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations

Paper • 2410.18860 • Published Oct 24, 2024 • 11

Analysing the Residual Stream of Language Models Under Knowledge Conflicts

Paper • 2410.16090 • Published Oct 21, 2024 • 7

Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering

Paper • 2410.15999 • Published Oct 21, 2024 • 20

Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2

Paper • 2408.05147 • Published Aug 9, 2024 • 40

upvoted an article 9 months ago

view article

Article

Improving Hugging Face Training Efficiency Through Packing with Flash Attention

Aug 21, 2024

• 34

upvoted a collection 10 months ago

🔍 Interpretability & Analysis of LMs

Collection

Outstanding research in LM interpretability and evaluation, summarized • 109 items • Updated 12 days ago • 100

upvoted an article 11 months ago

view article

Article

Introducing RWKV — An RNN with the advantages of a transformer

May 15, 2023

• 21

upvoted 3 papers 11 months ago

Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation

Paper • 2406.13663 • Published Jun 19, 2024 • 7

A Simple and Effective L_2 Norm-Based Strategy for KV Cache Compression

Paper • 2406.11430 • Published Jun 17, 2024 • 24

Are We Done with MMLU?

Paper • 2406.04127 • Published Jun 6, 2024 • 40

upvoted an article about 1 year ago

view article

Article

The Hallucinations Leaderboard, an Open Effort to Measure Hallucinations in Large Language Models

Jan 29, 2024

• 26