arXiv subjects

Pin-Yu Chen

Publications and source records attributed to Pin-Yu Chen.

At least 73 records · Page 4Linked to original sources

arXiv2025

CoP: Agentic Red-teaming for Large Language Models using Composition of Principles

Recent advances in Large Language Models (LLMs) have spurred transformative applications in various domains, ranging from open-source to proprietary LLMs. However, jailbreak attacks, which aim to break safety alignment and user compliance by tricking the target LLMs into answering harmful and risky responses, are becoming an urgent concern. The practice of red-teaming for LLMs is to proactively explore potential risks and error-prone instances before the release of frontier AI technology. This paper proposes an agentic workflow to automate and scale the red-teaming process of LLMs through the Composition-of-Principles (CoP) framework, where human users provide a set of red-teaming principles as instructions to an AI agent to automatically orchestrate effective red-teaming strategies and generate jailbreak prompts. Distinct from existing red-teaming methods, our CoP framework provides a unified and extensible framework to encompass and orchestrate human-provided red-teaming principles to enable the automated discovery of new red-teaming strategies. When tested against leading LLMs, CoP reveals unprecedented safety risks by finding novel jailbreak prompts and improving the best-known single-turn attack success rate by up to 19.0 times.

cs.AI↗

arXiv2025

Data-Driven Lipschitz Continuity: A Cost-Effective Approach to Improve Adversarial Robustness

As deep neural networks (DNNs) are increasingly deployed in sensitive applications, ensuring their security and robustness has become critical. A major threat to DNNs arises from adversarial attacks, where small input perturbations can lead to incorrect predictions. Recent advances in adversarial training improve robustness by incorporating additional examples from external datasets or generative models. However, these methods often incur high computational costs, limiting their practicality and hindering real-world deployment. In this paper, we propose a cost-efficient alternative based on Lipschitz continuity that achieves robustness comparable to models trained with extensive supplementary data. Unlike conventional adversarial training, our method requires only a single pass over the dataset without gradient estimation, making it highly efficient. Furthermore, our method can integrate seamlessly with existing adversarial training frameworks and enhances the robustness of models without requiring extra generative data. Experimental results show that our approach not only reduces computational overhead but also maintains or improves the defensive capabilities of robust neural networks. This work opens a promising direction for developing practical, scalable defenses against adversarial attacks.

cs.LG↗

arXiv2025

Adversarial Attack-Defense Co-Evolution for LLM Safety Alignment via Tree-Group Dual-Aware Search and Optimization

Large Language Models (LLMs) have developed rapidly in web services, delivering unprecedented capabilities while amplifying societal risks. Existing works tend to focus on either isolated jailbreak attacks or static defenses, neglecting the dynamic interplay between evolving threats and safeguards in real-world web contexts. To mitigate these challenges, we propose ACE-Safety (Adversarial Co-Evolution for LLM Safety), a novel framework that jointly optimize attack and defense models by seamlessly integrating two key innovative procedures: (1) Group-aware Strategy-guided Monte Carlo Tree Search (GS-MCTS), which efficiently explores jailbreak strategies to uncover vulnerabilities and generate diverse adversarial samples; (2) Adversarial Curriculum Tree-aware Group Policy Optimization (AC-TGPO), which jointly trains attack and defense LLMs with challenging samples via curriculum reinforcement learning, enabling robust mutual improvement. Evaluations across multiple benchmarks demonstrate that our method outperforms existing attack and defense approaches, and provides a feasible pathway for developing LLMs that can sustainably support responsible AI ecosystems.

cs.CR↗

arXiv2025

ICX360: In-Context eXplainability 360 Toolkit

Large Language Models (LLMs) have become ubiquitous in everyday life and are entering higher-stakes applications ranging from summarizing meeting transcripts to answering doctors' questions. As was the case with earlier predictive models, it is crucial that we develop tools for explaining the output of LLMs, be it a summary, list, response to a question, etc. With these needs in mind, we introduce In-Context Explainability 360 (ICX360), an open-source Python toolkit for explaining LLMs with a focus on the user-provided context (or prompts in general) that are fed to the LLMs. ICX360 contains implementations for three recent tools that explain LLMs using both black-box and white-box methods (via perturbations and gradients respectively). The toolkit, available at https://github.com/IBM/ICX360, contains quick-start guidance materials as well as detailed tutorials covering use cases such as retrieval augmented generation, natural language generation, and jailbreaking.

cs.CL↗

arXiv2025

VQC-MLPNet: An Unconventional Hybrid Quantum-Classical Architecture for Scalable and Robust Quantum Machine Learning

Variational quantum circuits (VQCs) hold promise for quantum machine learning but face challenges in expressivity, trainability, and noise resilience. We propose VQC-MLPNet, a hybrid architecture where a VQC generates the first-layer weights of a classical multilayer perceptron during training, while inference is performed entirely classically. This design preserves scalability, reduces quantum resource demands, and enables practical deployment. We provide a theoretical analysis based on statistical learning and neural tangent kernel theory, establishing explicit risk bounds and demonstrating improved expressivity and trainability compared to purely quantum or existing hybrid approaches. These theoretical insights demonstrate exponential improvements in representation capacity relative to quantum circuit depth and the number of qubits, providing clear computational advantages over standalone quantum circuits and existing hybrid quantum architectures. Empirical results on diverse datasets, including quantum-dot classification and genomic sequence analysis, show that VQC-MLPNet achieves high accuracy and robustness under realistic noise models, outperforming classical and quantum baselines while using significantly fewer trainable parameters.

quant-ph↗

arXiv2025

Enhancing Sentiment Classification with Machine Learning and Combinatorial Fusion

This paper presents a novel approach to sentiment classification using the application of Combinatorial Fusion Analysis (CFA) to integrate an ensemble of diverse machine learning models, achieving state-of-the-art accuracy on the IMDB sentiment analysis dataset of 97.072\%. CFA leverages the concept of cognitive diversity, which utilizes rank-score characteristic functions to quantify the dissimilarity between models and strategically combine their predictions. This is in contrast to the common process of scaling the size of individual models, and thus is comparatively efficient in computing resource use. Experimental results also indicate that CFA outperforms traditional ensemble methods by effectively computing and employing model diversity. The approach in this paper implements the combination of a transformer-based model of the RoBERTa architecture with traditional machine learning models, including Random Forest, SVM, and XGBoost.

cs.LG↗

arXiv2025

Neural Network Reprogrammability: A Unified Theme on Model Reprogramming, Prompt Tuning, and Prompt Instruction

As large-scale pre-trained foundation models continue to expand in size and capability, efficiently adapting them to specific downstream tasks has become increasingly critical. Despite substantial progress, existing adaptation approaches have evolved largely in isolation, without a clear understanding of their interrelationships. This survey introduces neural network reprogrammability as a unifying framework that bridges mainstream model adaptation techniques--model reprogramming, prompt tuning, and prompt instruction--previously fragmented research areas yet converges on a shared principle: repurposing a pre-trained model by manipulating information at the interfaces while keeping the model parameters frozen. These methods exploit neural networks' sensitivity to manipulation on different interfaces, be it through perturbing inputs, inserting tokens into intermediate layers, or providing task-specific examples in context, to redirect model behaviors towards desired outcomes. We then present a taxonomy that categorizes such information manipulation-based adaptation approaches across four key dimensions: manipulation format (fixed or learnable), location (interfaces where manipulations occur), operator (how they are applied), and output alignment requirement (post-processing needed to align outputs with downstream tasks). Notably, this framework applies consistently across data modalities, independent of specific model architectures. Moreover, viewing established techniques like in-context learning and chain-of-thought prompting through this lens reveals both their theoretical connections and practical distinctions. We further analyze remaining technical challenges and ethical considerations, positioning neural network reprogrammability as a fundamental paradigm for efficient model adaptation. We lastly identify promising research directions emerging from this integrative viewpoint.

cs.LG↗

arXiv2025

CarBoN: Calibrated Best-of-N Sampling Improves Test-time Reasoning

Allocating more computation during inference time (test-time scaling) improves language model performance, especially for reasoning tasks. However, popular methods like Best-of-$N$ sampling often show diminishing returns as $N$ increases. To address this inefficiency, we introduce a general test-time calibration framework that adaptively modifies the model toward high-reward reasoning paths, with theoretical guarantees of improving the lower bound of expected reward under finite sampling, all without large language model (LLM) retraining. Within this framework, we propose CarBoN (Calibrated Best-of-$N$), a two-phase method that first explores the solution space and then learns a calibration of the logits via an input-specific temperature $T$ and additive shift vector $δ$, guiding generation toward more reliable reasoning. Experiments on MATH-500 and AIME-2024 show that CarBoN improves efficiency, with up to $4\times$ fewer rollouts to reach the same accuracy, while often achieving higher accuracy under fixed budgets. We also analyze the complementary roles of $T$ and $δ$ in balancing output diversity and correctness, and demonstrate that the framework also generalizes to step-level sampling strategies such as beam search. For more information, please refer to our project page at huggingface.co/spaces/TrustSafeAI/Test-Time-Calibration.

cs.LG↗

arXiv2025

LLM Unlearning on Noisy Forget Sets: A Study of Incomplete, Rewritten, and Watermarked Data

Large language models (LLMs) exhibit remarkable generative capabilities but raise ethical and security concerns by memorizing sensitive data, reinforcing biases, and producing harmful content. These risks have spurred interest in LLM unlearning, the task of removing knowledge associated with undesirable data from pre-trained models. However, most existing methods assume access to clean, well-defined forget data samples, whereas real-world forget data could often be low-quality, synthetically rewritten, or watermarked, casting doubt on the reliability of unlearning. This work presents the first study of unlearning under perturbed or low-fidelity forget data, referred to as noisy forget sets. By systematically benchmarking state-of-the-art LLM unlearning methods, RMU and NPO, on such noisy forget sets, we find that unlearning remains surprisingly robust to perturbations, provided that core semantic signals are preserved. To explain this robustness, we propose a saliency-based interpretation: key semantic components that drive forgetting remain consistently influential despite substantial variation in surface form. This suggests that unlearning algorithms are primarily guided by deep semantic cues rather than shallow lexical patterns.

cs.LG↗

arXiv2025

Building a Foundational Guardrail for General Agentic Systems via Synthetic Data

While LLM agents can plan multi-step tasks, intervening at the planning stage-before any action is executed-is often the safest way to prevent harm, since certain risks can lead to severe consequences once carried out. However, existing guardrails mostly operate post-execution, which is difficult to scale and leaves little room for controllable supervision at the plan level. To address this challenge, we highlight three critical gaps in current research: data gap, model gap, and evaluation gap. To close the data gap, we introduce AuraGen, a controllable engine that (i) synthesizes benign trajectories, (ii) injects category-labeled risks with calibrated difficulty, and (iii) filters outputs via an automated reward model, producing large and reliable corpora for pre-execution safety. To close the guardian model gap, we propose a foundational guardrail Safiron, combining a cross-planner adapter with a compact guardian model. The adapter unifies different input formats, while Safiron flags risky cases, assigns risk types, and generates rationales; trained in two stages with a broadly explored data recipe, Safiron achieves robust transfer across settings. To close the evaluation gap, we release Pre-Exec Bench, a realistic benchmark covering diverse tools and branching trajectories, which measures detection, fine-grained categorization, explanation, and cross-planner generalization in human-verified scenarios. Extensive experiments demonstrate consistent gains of the proposed guardrail over strong baselines on Pre-Exec Bench, and ablations further distill actionable practices, providing a practical template for safer agentic systems.

cs.LG↗

arXiv2025

Adaptive Distraction: Probing LLM Contextual Robustness with Automated Tree Search

Large Language Models (LLMs) often struggle to maintain their original performance when faced with semantically coherent but task-irrelevant contextual information. Although prior studies have explored this issue using fixed-template or retrieval-based distractions, such static methods show limited effectiveness against contemporary models. To address this problem, we propose a dynamic distraction generation framework based on tree search, where the generation process is guided by model behavior. Without modifying the original question or answer, the method efficiently produces challenging adaptive distractions across multiple datasets, enabling systematic stress testing of LLMs' contextual robustness. Experiments on four benchmarks demonstrate that the generated distractions lead to an average performance drop of over 45\% for mainstream models. Further comparisons of mitigation strategies show that prompt-based optimization methods yield limited gains, whereas post-training approaches (e.g., DPO) significantly enhance the model's contextual robustness. The results indicate that these issues do not stem from knowledge deficits in LLMs, but from a fundamental inability to maintain consistent reasoning under contextual distraction, posing a major challenge to the reliability of LLMs in real-world applications. The code is publicly available at https://github.com/wyf23187/Adaptive_Distractions.

cs.CL↗

arXiv2025

Rethinking Data Protection in the (Generative) Artificial Intelligence Era

The (generative) artificial intelligence (AI) era has profoundly reshaped the meaning and value of data. No longer confined to static content, data now permeates every stage of the AI lifecycle from the training samples that shape model parameters to the prompts and outputs that drive real-world model deployment. This shift renders traditional notions of data protection insufficient, while the boundaries of what needs safeguarding remain poorly defined. Failing to safeguard data in AI systems can inflict societal and individual, underscoring the urgent need to clearly delineate the scope of and rigorously enforce data protection. In this perspective, we propose a four-level taxonomy, including non-usability, privacy preservation, traceability, and deletability, that captures the diverse protection needs arising in modern (generative) AI models and systems. Our framework offers a structured understanding of the trade-offs between data utility and control, spanning the entire AI pipeline, including training datasets, model weights, system prompts, and AI-generated content. We analyze representative technical approaches at each level and reveal regulatory blind spots that leave critical assets exposed. By offering a structured lens to align future AI technologies and governance with trustworthy data practices, we underscore the urgency of rethinking data protection for modern AI techniques and provide timely guidance for developers, researchers, and regulators alike.

cs.LG↗

arXiv2025

Bridging Models to Defend: A Population-Based Strategy for Robust Adversarial Defense

Adversarial robustness is a critical measure of a neural network's ability to withstand adversarial attacks at inference time. While robust training techniques have improved defenses against individual $\ell_p$-norm attacks (e.g., $\ell_2$ or $\ell_\infty$), models remain vulnerable to diversified $\ell_p$ perturbations. To address this challenge, we propose a novel Robust Mode Connectivity (RMC)-oriented adversarial defense framework comprising two population-based learning phases. In Phase I, RMC searches the parameter space between two pre-trained models to construct a continuous path containing models with high robustness against multiple $\ell_p$ attacks. To improve efficiency, we introduce a Self-Robust Mode Connectivity (SRMC) module that accelerates endpoint generation in RMC. Building on RMC, Phase II presents RMC-based optimization, where RMC modules are composed to further enhance diversified robustness. To increase Phase II efficiency, we propose Efficient Robust Mode Connectivity (ERMC), which leverages $\ell_1$- and $\ell_\infty$-adversarially trained models to achieve robustness across a broad range of $p$-norms. An ensemble strategy is employed to further boost ERMC's performance. Extensive experiments across diverse datasets and architectures demonstrate that our methods significantly improve robustness against $\ell_\infty$, $\ell_2$, $\ell_1$, and hybrid attacks. Code is available at https://github.com/wangren09/MCGR.

cs.AI↗

arXiv2025

Attention Slipping: A Mechanistic Understanding of Jailbreak Attacks and Defenses in LLMs

As large language models (LLMs) become more integral to society and technology, ensuring their safety becomes essential. Jailbreak attacks exploit vulnerabilities to bypass safety guardrails, posing a significant threat. However, the mechanisms enabling these attacks are not well understood. In this paper, we reveal a universal phenomenon that occurs during jailbreak attacks: Attention Slipping. During this phenomenon, the model gradually reduces the attention it allocates to unsafe requests in a user query during the attack process, ultimately causing a jailbreak. We show Attention Slipping is consistent across various jailbreak methods, including gradient-based token replacement, prompt-level template refinement, and in-context learning. Additionally, we evaluate two defenses based on query perturbation, Token Highlighter and SmoothLLM, and find they indirectly mitigate Attention Slipping, with their effectiveness positively correlated with the degree of mitigation achieved. Inspired by this finding, we propose Attention Sharpening, a new defense that directly counters Attention Slipping by sharpening the attention score distribution using temperature scaling. Experiments on four leading LLMs (Gemma2-9B-It, Llama3.1-8B-It, Qwen2.5-7B-It, Mistral-7B-It v0.2) show that our method effectively resists various jailbreak attacks while maintaining performance on benign tasks on AlpacaEval. Importantly, Attention Sharpening introduces no additional computational or memory overhead, making it an efficient and practical solution for real-world deployment.

cs.CR↗

arXiv2025

Graph is all you need? Lightweight data-agnostic neural architecture search without training

Neural architecture search (NAS) enables the automatic design of neural network models. However, training the candidates generated by the search algorithm for performance evaluation incurs considerable computational overhead. Our method, dubbed nasgraph, remarkably reduces the computational costs by converting neural architectures to graphs and using the average degree, a graph measure, as the proxy in lieu of the evaluation metric. Our training-free NAS method is data-agnostic and light-weight. It can find the best architecture among 200 randomly sampled architectures from NAS-Bench201 in 217 CPU seconds. Besides, our method is able to achieve competitive performance on various datasets including NASBench-101, NASBench-201, and NDS search spaces. We also demonstrate that nasgraph generalizes to more challenging tasks on Micro TransNAS-Bench-101.

cs.LG↗

arXiv2025

Optimization-Free Universal Watermark Forgery with Regenerative Diffusion Models

Watermarking becomes one of the pivotal solutions to trace and verify the origin of synthetic images generated by artificial intelligence models, but it is not free of risks. Recent studies demonstrate the capability to forge watermarks from a target image onto cover images via adversarial optimization without knowledge of the target generative model and watermark schemes. In this paper, we uncover a greater risk of an optimization-free and universal watermark forgery that harnesses existing regenerative diffusion models. Our proposed forgery attack, PnP (Plug-and-Plant), seamlessly extracts and integrates the target watermark via regenerating the image, without needing any additional optimization routine. It allows for universal watermark forgery that works independently of the target image's origin or the watermarking model used. We explore the watermarked latent extracted from the target image and visual-textual context of cover images as priors to guide sampling of the regenerative process. Extensive evaluation on 24 scenarios of model-data-watermark combinations demonstrates that PnP can successfully forge the watermark (up to 100% detectability and user attribution), and maintain the best visual perception. By bypassing model retraining and enabling adaptability to any image, our approach significantly broadens the scope of forgery attacks, presenting a greater challenge to the security of current watermarking techniques for diffusion models and the authority of watermarking schemes in synthetic data generation and governance.

cs.MM↗

arXiv2025

Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets

Recent advancements in large language models (LLMs) have underscored their vulnerability to safety alignment jailbreaks, particularly when subjected to downstream fine-tuning. However, existing mitigation strategies primarily focus on reactively addressing jailbreak incidents after safety guardrails have been compromised, removing harmful gradients during fine-tuning, or continuously reinforcing safety alignment throughout fine-tuning. As such, they tend to overlook a critical upstream factor: the role of the original safety-alignment data. This paper therefore investigates the degradation of safety guardrails through the lens of representation similarity between upstream alignment datasets and downstream fine-tuning tasks. Our experiments demonstrate that high similarity between these datasets significantly weakens safety guardrails, making models more susceptible to jailbreaks. Conversely, low similarity between these two types of datasets yields substantially more robust models and thus reduces harmfulness score by up to 10.33%. By highlighting the importance of upstream dataset design in the building of durable safety guardrails and reducing real-world vulnerability to jailbreak attacks, these findings offer actionable insights for fine-tuning service providers.

cs.CR↗

arXiv2025

Defensive Prompt Patch: A Robust and Interpretable Defense of LLMs against Jailbreak Attacks

Safety, security, and compliance are essential requirements when aligning large language models (LLMs). However, many seemingly aligned LLMs are soon shown to be susceptible to jailbreak attacks. These attacks aim to circumvent the models' safety guardrails and security mechanisms by introducing jailbreak prompts into malicious queries. In response to these challenges, this paper introduces Defensive Prompt Patch (DPP), a novel prompt-based defense mechanism specifically designed to protect LLMs against such sophisticated jailbreak strategies. Unlike previous approaches, which have often compromised the utility of the model for the sake of safety, DPP is designed to achieve a minimal Attack Success Rate (ASR) while preserving the high utility of LLMs. Our method uses strategically designed interpretable suffix prompts that effectively thwart a wide range of standard and adaptive jailbreak techniques. Empirical results conducted on LLAMA-2-7B-Chat and Mistral-7B-Instruct-v0.2 models demonstrate the robustness and adaptability of DPP, showing significant reductions in ASR with negligible impact on utility. Our approach not only outperforms existing defense strategies in balancing safety and functionality, but also provides a scalable and interpretable solution applicable to various LLM platforms.

cs.CR↗