CoreInfer: Accelerating Large Language Model Inference with Semantics-Inspired Adaptive Sparse Activation

TL;DR

CoreInfer introduces a sentence-level, semantics-inspired adaptive sparse activation method for large language models, significantly accelerating inference by predicting core neurons without additional MLP computations.

Contribution

It proposes a novel MLP-free, sentence-wise core neuron prediction approach based on semantic stability, enabling zero-cost sparse inference for LLMs.

Findings

Achieved 10.33x speedup over Huggingface implementation.

Demonstrated effectiveness across various models and tasks.

Core neurons show stability and semantic similarity, validating the approach.

Abstract

Large language models (LLMs) with billions of parameters have sparked a new wave of exciting AI applications. However, their high computational costs and memory demands during inference pose significant challenges. Adaptive sparse activation inference, which activates only a small number of neurons for each token, offers a novel way to accelerate model inference without degrading performance, showing great potential for resource-constrained hardware devices. Nevertheless, existing methods predict activated neurons based on individual tokens with additional MLP, which involve frequent changes in activation maps and resource calls, limiting the acceleration benefits of sparse activation. In this paper, we introduce CoreInfer, an MLP-free adaptive sparse activation inference method based on sentence-level prediction. Specifically, we propose the concept of sentence-wise core neurons, which refers to the subset of neurons most critical for a given sentence, and empirically demonstrate its effectiveness. To determine the core neurons, we explore the correlation between core neurons and the sentence's semantics. Remarkably, we discovered that core neurons exhibit both stability and similarity in relation to the sentence's semantics -- an insight overlooked by previous studies. Building on this finding, we further design two semantic-based methods for predicting core neurons to fit different input scenarios. In CoreInfer, the core neurons are determined during the pre-filling stage and fixed during the encoding stage, enabling zero-cost sparse inference. We evaluated the model generalization and task generalization of CoreInfer across various models and tasks. Notably, on an NVIDIA TITAN XP GPU, CoreInfer achieved a 10.33 times and 2.72 times speedup compared to the Huggingface implementation and PowerInfer, respectively.