paper-with-me

Papers

Leveraging Speculative Sampling and KV-Cache Optimizations Together for Generative AI using OpenVINO

2023-11-08 · Haim Barad, Ekaterina Aidova, Yury Gorbachev

Inference optimizations are critical for improving user experience and reducing infrastructure costs and power consumption. In this article, we illustrate a form of dynamic execution known as speculative sampling to reduce the overall latency of text generation and compare it with standard autoregressive sampling. This can be used together with model-based optimizations (e.g. quantization) to provide an optimized solution. Both sampling methods make use of KV caching. A Jupyter notebook and some sample executions are provided.

📄 PDF Abstract BibTeX arXiv:2311.04951

Code (1)

openvinotoolkit/openvino_notebooks 공식 구현 pytorch

Tasks

QuantizationText Generation

Similar Papers 제목 키워드 기반

Learning Harmonized Representations for Speculative Sampling

2024-08-28 · Lefan Zhang, Xiaodan Wang, Yanhua Huang, Ruiwen Xu

Speculative sampling is a promising approach to accelerate the decoding stage for Large Language Models (LLMs). Recent advancements that leverage target LLM's contextual information, such as hidden states and KV cache, h…

Optimizing Agentic Language Model Inference via Speculative Tool Calls

2025-12-17 · Daniel Nichols, Prajwal Singhania, Charles Jekel, Abhinav Bhatele 외 arxiv

Language models (LMs) are becoming increasingly dependent on external tools. LM-based agentic frameworks frequently interact with their environment via such tools to search files, run code, call APIs, etc. Further, moder…

CARD: A Cache-Assisted Parallel Speculative Decoding Framework via Query-and-Correct Paradigm for Accelerating LLM Inference

2025-08-06 · Enyu Zhou, Kai Sheng, Hao Chen, Xin He arxiv

Speculative decoding (SD), where a draft model provides multiple candidate tokens for the target model to verify in parallel, has demonstrated significant potential for accelerating LLM inference. Yet, existing SD approa…

Cacheback: Speculative Decoding With Nothing But Cache

2025-11-15 · Zhiyao Ma, In Gim, Lin Zhong arxiv

We present Cacheback Decoding, a training-free and model-agnostic speculative decoding method that exploits the locality in language to accelerate Large Language Model (LLM) inference. Cacheback leverages only Least Rece…

SpecAttn: Speculating Sparse Attention

2025-10-31 · Harsh Shah arxiv

Large Language Models (LLMs) face significant computational bottlenecks during inference due to the quadratic complexity of self-attention mechanisms, particularly as context lengths increase. We introduce SpecAttn, a no…