4 papers
Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving
Rui Li, Zhaoning Zhang, Libo Zhang +3
Speculative decoding (SD) accelerates LLM inference by verifying draft tokens in parallel. However, this method presents a critical trade-off: it improves throughput in low-load, m…
A Digital SRAM-Based Compute-In-Memory Macro for Weight-Stationary Dynamic Matrix Multiplication in Transformer Attention Score Computation
Jianyi Yu, Tengxiao Wang, Yuxuan Wang +6
Compute-in-memory (CIM) techniques are widely employed in energy-efficient artificial intelligent (AI) processors. They alleviate power and latency bottlenecks caused by extensive…
MIV: Towards Efficient and Fine-grained Multimodal In-Context Learning via Representation Engineering
Yanshu Li, Yi Cao, Hongyang He +5
Multimodal in-context learning (ICL) equips Large Vision-language Models (LVLMs) with the ability to adapt to new tasks via multiple user-provided demonstrations, without requiring…
Can an Easy-to-Hard Curriculum Make Reasoning Emerge in Small Language Models? Evidence from a Four-Stage Curriculum on GPT-2
Xiang Fu
We demonstrate that a developmentally ordered curriculum markedly improves reasoning transparency and sample-efficiency in small language models (SLMs). Concretely, we train Cogniv…