-
CatLIP: CLIP-level Visual Recognition Accuracy with 2.7x Faster Pre-training on Web-scale Image-Text Data
Paper • 2404.15653 • Published • 27 -
MoDE: CLIP Data Experts via Clustering
Paper • 2404.16030 • Published • 13 -
MoRA: High-Rank Updating for Parameter-Efficient Fine-Tuning
Paper • 2405.12130 • Published • 47 -
Reducing Transformer Key-Value Cache Size with Cross-Layer Attention
Paper • 2405.12981 • Published • 29
Collections
Discover the best community collections!
Collections including paper arxiv:2410.10814
-
RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval
Paper • 2401.18059 • Published • 37 -
Personalized Visual Instruction Tuning
Paper • 2410.07113 • Published • 70 -
Differential Transformer
Paper • 2410.05258 • Published • 169 -
What Matters in Transformers? Not All Attention is Needed
Paper • 2406.15786 • Published • 30
-
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
Paper • 2309.14509 • Published • 17 -
LLM Augmented LLMs: Expanding Capabilities through Composition
Paper • 2401.02412 • Published • 37 -
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
Paper • 2401.06066 • Published • 47 -
Tuning Language Models by Proxy
Paper • 2401.08565 • Published • 22
-
LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models
Paper • 2309.12307 • Published • 88 -
Qwen Technical Report
Paper • 2309.16609 • Published • 35 -
RLHF Workflow: From Reward Modeling to Online RLHF
Paper • 2405.07863 • Published • 67 -
Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free
Paper • 2410.10814 • Published • 49
-
Chain-of-Verification Reduces Hallucination in Large Language Models
Paper • 2309.11495 • Published • 37 -
Adapting Large Language Models via Reading Comprehension
Paper • 2309.09530 • Published • 77 -
CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages
Paper • 2309.09400 • Published • 84 -
Language Modeling Is Compression
Paper • 2309.10668 • Published • 83