QLoRA: Efficient Finetuning of Quantized LLMs
Paper
• 2305.14314 • Published
• 59
llm-performance
-
-
Training Transformers with 4-bit Integers
Paper • 2306.11987 • Published • 23 -
FasterViT: Fast Vision Transformers with Hierarchical Attention
Paper • 2306.06189 • Published • 32 -
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
Paper • 2309.14509 • Published • 20 -
VeRA: Vector-based Random Matrix Adaptation
Paper • 2310.11454 • Published • 30 -
LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Models
Paper • 2310.08659 • Published • 27 -
Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Paper • 2310.17157 • Published • 14 -
TinyGPT-V: Efficient Multimodal Large Language Model via Small Backbones
Paper • 2312.16862 • Published • 31 -
SparQ Attention: Bandwidth-Efficient LLM Inference
Paper • 2312.04985 • Published • 40 -
PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU
Paper • 2312.12456 • Published • 45 -
LLM in a flash: Efficient Large Language Model Inference with Limited Memory
Paper • 2312.11514 • Published • 260 -
LLM Augmented LLMs: Expanding Capabilities through Composition
Paper • 2401.02412 • Published • 38 -
SliceGPT: Compress Large Language Models by Deleting Rows and Columns
Paper • 2401.15024 • Published • 73 -
Step-3 is Large yet Affordable: Model-system Co-design for Cost-effective Decoding
Paper • 2507.19427 • Published • 21