HiFuzz: Hierarchical Reinforcement Learning for Semantic-Aware and Adaptive CPU Fuzzing

Paper

First seen 7/10/2026

Last seen 7/10/2026

Evidence 2 chunks

NEIGHBORHOOD

No graph connections found for this entity yet. It may appear in future ingestion runs.

11 connections

HiFuzz introduces → 100% 1e

The paper introduces HiFuzz as a novel hierarchical reinforcement learning framework for CPU fuzzing.

Ya Wang authored by → 100% 1e

Ya Wang is listed as an author of the HiFuzz paper.

Hanwei Fan authored by → 100% 1e

Hanwei Fan is listed as an author of the HiFuzz paper.

Zhenguo Liu authored by → 100% 1e

Zhenguo Liu is listed as an author of the HiFuzz paper.

Xiaofeng Zhou authored by → 100% 1e

Xiaofeng Zhou is listed as an author of the HiFuzz paper.

Yangdi Lyu authored by → 100% 1e

Yangdi Lyu is listed as an author of the HiFuzz paper.

Jiang Xu authored by → 100% 1e

Jiang Xu is listed as an author of the HiFuzz paper.

Wei Zhang authored by → 100% 1e

Wei Zhang is listed as an author of the HiFuzz paper.

The paper mentions speculative execution vulnerabilities to motivate the need for better hardware verification.

Moore's Law mentions → 95% 1e

The paper mentions the diminishing returns of Moore's Law as context for increased microarchitectural complexity.

Reward Sparsity mentions → 100% 1e

The paper identifies reward sparsity as a key challenge for RL-based hardware fuzzing.