Benchmarking On-Device Machine Learning on Apple Silicon with MLX
"推理延迟" 相关文章
ZSPAPrune: Zero-Shot Prompt-Aware Token Pruning for Vision-Language Models
BeLLMan: Controlling LLM Congestion
BeLLMan: Controlling LLM Congestion
BILLY: Steering Large Language Models via Merging Persona Vectors for Creative Generation
Adaptive Dual Reasoner: Large Reasoning Models Can Think Efficiently by Hybrid Reasoning
Taming Latency-Memory Trade-Off in MoE-Based LLM Serving via Fine-Grained Expert Offloading
Elastic On-Device LLM Service
Nav-EE: Navigation-Guided Early Exiting for Efficient Vision-Language Models in Autonomous Driving
Nav-EE: Navigation-Guided Early Exiting for Efficient Vision-Language Models in Autonomous Driving
From Long to Lean: Performance-aware and Adaptive Chain-of-Thought Compression via Multi-round Refinement
Learning Primitive Embodied World Models: Towards Scalable Robotic Learning
Slim-SC: Thought Pruning for Efficient Scaling with Self-Consistency
Accelerating Mixture-of-Expert Inference with Adaptive Expert Split Mechanism
SpecDec++: Boosting Speculative Decoding via Adaptive Candidate Lengths
