mlsanityData to 14 Oct 2026

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

arXiv 2607.06619 ↗First posted 7 Jul 2026cs.AR cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL training of language and agent models

Abstract (arXiv)

Modern processor verification struggles to reach deep architectural states due to the inefficiencies of traditional mutation-based fuzzing. We propose HiFuzz, a novel hierarchical reinforcement learning framework that replaces mutation with a structured, two-layer generation process: a Program Agent for global layout and a Basic Block Agent for precise instruction filling. To overcome reward sparsity, HiFuzz integrates an adaptive coverage reward mechanism and a semantic-aware basic block encoder providing intrinsic feedback. Extensive evaluations on three real-world RISC-V cores demonstrate that HiFuzz significantly outperforms state-of-the-art fuzzers in coverage and bug detection.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL training of language and agent models, one of 15 directions in Deep reinforcement learning methods. Builds on 7 of its 24 references that are in the map. First posted on arXiv 7 Jul 2026.

Builds on

7 of 24 references.

References · 7

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  3. 2018Exploration by Random Network Distillationcited by 993
  4. 2017Rainbow: Combining Improvements in Deep Reinforcement Learningcited by 786
  5. 2018Spectre Attacks: Exploiting Speculative Executioncited by 47
  6. 2019The Cost of Application-Class Processing: Energy and Performance Analysis of a Linux-ready 1.7GHz 64bit RISC-V Core in 22nm FDSOI Technologycited by 17
  7. 2024Beyond Random Inputs: A Novel ML-Based Hardware Fuzzingcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025μRL: Discovering Transient Execution Vulnerabilities Using Reinforcement Learning
  2. 2026EvolveGen: Algorithmic Level Hardware Model Checking Benchmark Generation through Reinforcement Learning
  3. 2025ChipSeek: Optimizing Verilog Generation via EDA-Integrated Reinforcement Learning
  4. 2026BEHAVE: Functional Behavior Modeling Enables Self-Improving Agents for Hardware Design and Verification
  5. 2025SynFuzz: Leveraging Fuzzing of Netlist to Detect Synthesis Bugs ↗
  6. 2025Towards Reinforcement Learning for Exploration of Speculative Execution Vulnerabilities
  7. 2025DAPO: Design Structure-Aware Pass Ordering in High-Level Synthesis with Graph Contrastive and Reinforcement Learning
  8. 2026HLS-Seek: QoR-Aware Code Generation for High-Level Synthesis via Proxy Comparative Reward Reinforcement Learning