Ping Guo

Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-Based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions.

The discovery of symbolic solutions—mathematical expressions, logical rules, and algorithmic structures—is fundamental to advancing scientific and engineering progress. However, traditional methods often struggle with search efficiency and fail to integrate knowledge effectively. While recent large language model-based (LLM-based) approaches have demonstrated improvements in search efficiency, they lack the ability to continually refine and expand upon discovered solutions and their underlying knowledge, limiting their potential for \textitopen-ended innovation. To address these limitations, we introduce CoEvo, a novel framework that leverages large language models within an evolutionary search methodology to continually generate and refine symbolic solutions. CoEvo integrates a dynamic knowledge library, enabling open-ended innovation of solutions through effective knowledge management. Additionally, CoEvo leverages multiple representations of solutions—including natural language, mathematical expressions, and code—to further enhance search efficiency. By combining the reasoning capabilities of LLMs with the exploratory power of evolutionary algorithms, CoEvo significantly improves the efficiency and scope of symbolic discovery. Our experimental results demonstrate that this method not only enhances the efficiency of searching for symbolic solutions but also supports the ongoing discovery process, akin to human scientific endeavors. This study represents a first effort in conceptualizing the search for symbolic solutions as a lifelong, iterative process, marking a significant step towards harnessing LLMs in the perpetual pursuit of scientific and engineering breakthroughs. Our code is available at \urlhttps://github.com/pgg3/CoEvo.

Optimizing Register Transfer Level (RTL) code is crucial for improving the efficiency and performance of digital circuits in the early stages of synthesis. Manual rewriting, guided by synthesis feedback, can yield high-quality results but is time-consuming and error-prone. Most existing compiler-based approaches have difficulty handling complex design constraints. Large Language Model (LLM)-based methods have emerged as a promising alternative to address these challenges. However, LLM-based approaches often face difficulties in ensuring alignment between the generated code and the provided prompts. This paper introduces SymRTLO, a neuron-symbolic framework that integrates LLMs with symbolic reasoning for the efficient and effective optimization of RTL code. Our method incorporates a retrieval-augmented system of optimization rules and Abstract Syntax Tree (AST)-based templates, enabling LLM-based rewriting that maintains syntactic correctness while minimizing undesired circuit behaviors. A symbolic module is proposed for analyzing and optimizing finite state machine (FSM) logic, allowing fine-grained state merging and partial specification handling beyond the scope of pattern-based compilers. Furthermore, an efficient verification pipeline, combining formal equivalence checks with test-driven validation, further reduces the complexity of verification. Experiments on the RTL-Rewriter benchmark with Synopsys Design Compiler and Yosys show that SymRTLO improves power, performance, and area (PPA) by up to 43.9%, 62.5%, and 51.1%, respectively, compared to the state-of-the-art methods. Our code is available at \urlhttps://github.com/NellyW8/SymRTLO

The GPS has empowered billions of users and various critical infrastructures with its positioning and time services. However, GPS spoofing attacks also become a growing threat to GPS-dependent systems. Existing detection methods either require expensive hardware modifications to current GPS devices or lack the basic robustness against sophisticated attacks, hurting their adoption and usage in practice. In this paper, we propose a novel GPS spoofing detection framework that works with off-the-shelf GPS chipsets. Our basic idea is to rotate a one-side-blocked GPS receiver to derive the angle-of-arrival (AoAs) of received signals and compare them with the GPS constellation (consists of tens of GPS satellites). We first demonstrate the effectiveness of this idea by implementing a smartphone prototype and evaluating it against a real spoofer in various field experiments (in both open air and urban canyon environments). Our method achieves a high accuracy (95%–100%) in 5 seconds. Then we implement an adaptive attack, assuming the attacker becomes aware of our defense method and actively modulates the spoofing signals accordingly. We study this adaptive attack and propose enhancement methods (using the rotation speed as the "secret key") to fortify the defense. Further experiments are conducted to validate the effectiveness of the enhanced defense.

The escalating threat of adversarial attacks on deep learning models, particularly in security-critical fields, has underscored the need for robust deep learning systems. Conventional robustness evaluations have relied on adversarial accuracy, which measures a model’s performance under a specific perturbation intensity. However, this singular metric does not fully encapsulate the overall resilience of a model against varying degrees of perturbation. To address this gap, we propose a new metric termed adversarial hypervolume, assessing the robustness of deep learning models comprehensively over a range of perturbation intensities from a multi-objective optimization standpoint. This metric allows for an in-depth comparison of defense mechanisms and recognizes the trivial improvements in robustness afforded by less potent defensive strategies. Additionally, we adopt a novel training algorithm that enhances adversarial robustness uniformly across various perturbation intensities, in contrast to methods narrowly focused on optimizing adversarial accuracy. Our extensive empirical studies validate the effectiveness of the adversarial hypervolume metric, demonstrating its ability to reveal subtle differences in robustness that adversarial accuracy overlooks. This research contributes a new measure of robustness and establishes a standard for assessing and benchmarking the resilience of current and future defensive models against adversarial threats.

Hypervolume subset selection (HSS) has received significant attention since it has a strong connection with evolutionary multi-objective optimization (EMO), such as environment selection and post-processing to identify representative solutions for decision-makers. The goal of HSS is to find the optimal subset that maximizes the hypervolume indicator subject to a given cardinality constraint. However, existing HSS algorithms or related methods are not efficient in achieving good performance in high-dimensional objective spaces. This is primarily because HSS problems become NP-hard when the number of objectives exceeds two, and the calculation of hypervolume contribution is very time-consuming. To efficiently solve HSS problems while maintaining a good solution quality, we propose a fast and scalable hypervolume subset selection method for many-objective optimization based on the determinantal point process (DPP), named DPP-HSS, which is fully free of hypervolume contribution calculation. Specifically, DPP-HSS constructs a hypervolume kernel matrix by extracting the convergence and diversity representations of each solution for a given HSS problem. This matrix is then used to build a DPP model. Subsequently, the original HSS problem is reformulated as a new maximization optimization problem based on the constructed model. A greedy DPP-based hypervolume subset selection algorithm is implemented to solve this transformed problem. Extensive experiments show that the proposed DPP-HSS achieves significant speedup and good hypervolume performance in comparison with state-of-the-art HSS algorithms on benchmark problems. Furthermore, DPP-HSS demonstrates very good scalability with respect to the number of objectives.

Large Language Models (LLMs) have demonstrated great potential in automating the generation of Verilog hardware description language code for hardware design. This automation is critical to reducing human effort in the complex and error-prone process of hardware design. However, existing approaches predominantly rely on human intervention and fine-tuning using curated datasets, limiting their scalability in automated design workflows. Although recent iterative search techniques have emerged, they often fail to explore diverse design solutions and may underperform simpler approaches such as repeated prompting. To address these limitations, we introduce EvoVerilog, a novel framework that combines the reasoning capabilities of LLMs with evolutionary algorithms to automatically generate and refine Verilog code. EvoVerilog utilizes a multiobjective, population-based search strategy to explore a wide range of design possibilities without requiring human intervention. Extensive experiments demonstrate that EvoVerilog achieves state-of-the-art performance, with pass@10 scores of 89.1 and 80.2 on the VerilogEval-Machine and VerilogEval-Human benchmarks, respectively. Furthermore, the framework showcases its ability to explore diverse designs by simultaneously generating a variety of functional Verilog code while optimizing resource utilization.

CUDA kernel optimization has become a critical bottleneck for AI performance, as deep learning training and inference efficiency directly depends on highly optimized GPU kernels. Despite the promise of Large Language Models (LLMs) for automating kernel optimization, this field suffers from a fragmented ecosystem of isolated and incomparable approaches with unclear problem formulations. Furthermore, general-purpose LLM code evolution methods cannot meet strict correctness requirements of CUDA kernel optimization. We address these fundamental challenges by first formalizing CUDA kernel optimization as a code optimization task with a clear objective, constraints, and evaluation metrics. We then establish the first systematic LLM-based code evolution framework, EvoEngineer, that provides guidance for designing and adapting optimization strategies to achieve a balance between performance and correctness. Finally, we implement a kernel optimization system based on this framework and conduct extensive experiments on 91 real-world CUDA kernels. Our results demonstrate that EvoEngineer achieves a principled balance between performance and correctness, with the highest averaged median speedup of \textbf2.72\times over baseline CUDA kernels and a code validity rate of \textbf69.8%, outperforming existing methods on both dimensions. Our method achieves a maximum speedup of \textbf36.75\times among all operations over PyTorch kernels and delivers the highest speedup on \textbf28 (\textbf56.0%) of 50 operations that achieve over \textbf2\times acceleration.