DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
Paper
•
2501.12948
•
Published
•
347
llm_improvement_research
-
-
LightThinker: Thinking Step-by-Step Compression
Paper • 2502.15589 • Published • 26 -
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
Paper • 2405.04434 • Published • 18 -
Model Compression and Efficient Inference for Large Language Models: A Survey
Paper • 2402.09748 • Published • 1 -
Efficient Transformers: A Survey
Paper • 2009.06732 • Published • 1 -
Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference
Paper • 2403.09054 • Published • 1 -
FastCache: Optimizing Multimodal LLM Serving through Lightweight KV-Cache Compression Framework
Paper • 2503.08461 • Published -
Beyond RAG: Task-Aware KV Cache Compression for Comprehensive Knowledge Reasoning
Paper • 2503.04973 • Published • 20