Search

The rapid advancement of generative models has introduced serious risks, including deepfake techniques for facial synthesis and editing. Traditional approaches rely on training classifiers and enhancing generalizability through various feature extraction techniques. Meanwhile, training-free detection methods address issues like limited data and overfitting by directly leveraging statistical properties from vision foundation models to distinguish between real and fake images. The current leading training-free approach, RIGID, utilizes DINOv2 sensitivity to perturbations in image space for detecting fake images, with fake image embeddings exhibiting greater sensitivity than those of real images. This observation prompts us to investigate how detection performance varies across model backbones, perturbation types, and datasets. Our experiments reveal that detection performance is closely linked to model robustness, with self-supervised (SSL) models providing more reliable representations. While Gaussian noise effectively detects general objects, it performs worse on facial images, whereas Gaussian blur is more effective due to potential frequency artifacts. To further improve detection, we introduce Contrastive Blur, which enhances performance on facial images, and MINDER (MINimum distance DetEctoR), which addresses noise type bias, balancing performance across domains. Beyond performance gains, our work offers valuable insights for both the generative and detection communities, contributing to a deeper understanding of model robustness property utilized for deepfake detection.

This paper introduces a hybrid numerical scheme for the fuzzy dark matter model: It combines a wave-based approach to solve the Schr\"odinger equation using Fourier continuations with Gram polynomials and a fluid-based approach to solve the Hamilton-Jacobi-Madelung equations. This hybrid scheme facilitates zoom-in simulations for cosmological volumes beyond the capabilities of wave-based solvers alone and accurately simulates the full nonlinear dynamics of fuzzy dark matter. We detail the implementation of a Hamilton-Jacobi-Madelung solver, the methodology for phase matching at fluid-wave boundaries, the development of a local pseudospectral wave solver based on Fourier continuations, new grid refinement criteria for both fluid and wave solvers, an interpolation algorithm based on Fourier continuations, and the integration of these building blocks into the adaptive mesh refinement code GAMER. The superiority of the scheme is demonstrated through various performance and accuracy tests, tracking the linear power spectrum evolution in a 10 Mpc/h box, and a hybrid cosmological simulation in a 5.6 Mpc/h box. The corresponding code is published as part of the GAMER project on https://github.com/gamer-project/gamer., Comment: 25 pages, 18 figures, submitted to ApJS

Text-to-image (T2I) models have shown remarkable progress, but their potential to generate harmful content remains a critical concern in the ML community. While various safety mechanisms have been developed, the field lacks systematic tools for evaluating their effectiveness against real-world misuse scenarios. In this work, we propose ICER, a novel red-teaming framework that leverages Large Language Models (LLMs) and a bandit optimization-based algorithm to generate interpretable and semantic meaningful problematic prompts by learning from past successful red-teaming attempts. Our ICER efficiently probes safety mechanisms across different T2I models without requiring internal access or additional training, making it broadly applicable to deployed systems. Through extensive experiments, we demonstrate that ICER significantly outperforms existing prompt attack methods in identifying model vulnerabilities while maintaining high semantic similarity with intended content. By uncovering that successful jailbreaking instances can systematically facilitate the discovery of new vulnerabilities, our work provides crucial insights for developing more robust safety mechanisms in T2I systems.

Quantum machine learning (QML) is a rapidly growing field that combines quantum computing principles with traditional machine learning. It seeks to revolutionize machine learning by harnessing the unique capabilities of quantum mechanics and employs machine learning techniques to advance quantum computing research. This paper introduces quantum computing for the machine learning paradigm, where variational quantum circuits (VQC) are used to develop QML architectures on noisy intermediate-scale quantum (NISQ) devices. We discuss machine learning for the quantum computing paradigm, showcasing our recent theoretical and empirical findings. In particular, we delve into future directions for studying QML, exploring the potential industrial impacts of QML research., Comment: In submission

Quantum Machine Learning (QML) offers tremendous potential but is currently limited by the availability of qubits. We introduce an innovative approach that utilizes pre-trained neural networks to enhance Variational Quantum Circuits (VQC). This technique effectively separates approximation error from qubit count and removes the need for restrictive conditions, making QML more viable for real-world applications. Our method significantly improves parameter optimization for VQC while delivering notable gains in representation and generalization capabilities, as evidenced by rigorous theoretical analysis and extensive empirical testing on quantum dot classification tasks. Moreover, our results extend to applications such as human genome analysis, demonstrating the broad applicability of our approach. By addressing the constraints of current quantum hardware, our work paves the way for a new era of advanced QML applications, unlocking the full potential of quantum computing in fields such as machine learning, materials science, medicine, mimetics, and various interdisciplinary areas., Comment: In submission

There is a growing interest in training domain-expert LLMs that excel in specific technical fields compared to their general-purpose instruction-tuned counterparts. However, these expert models often experience a loss in their safety abilities in the process, making them capable of generating harmful content. As a solution, we introduce an efficient and effective merging-based alignment method called \textsc{MergeAlign} that interpolates the domain and alignment vectors, creating safer domain-specific models while preserving their utility. We apply \textsc{MergeAlign} on Llama3 variants that are experts in medicine and finance, obtaining substantial alignment improvements with minimal to no degradation on domain-specific benchmarks. We study the impact of model merging through model similarity metrics and contributions of individual models being merged. We hope our findings open new research avenues and inspire more efficient development of safe expert LLMs.

Large Language Models (LLMs) are increasingly used to control robotic systems such as drones, but their risks of causing physical threats and harm in real-world applications remain unexplored. Our study addresses the critical gap in evaluating LLM physical safety by developing a comprehensive benchmark for drone control. We classify the physical safety risks of drones into four categories: (1) human-targeted threats, (2) object-targeted threats, (3) infrastructure attacks, and (4) regulatory violations. Our evaluation of mainstream LLMs reveals an undesirable trade-off between utility and safety, with models that excel in code generation often performing poorly in crucial safety aspects. Furthermore, while incorporating advanced prompt engineering techniques such as In-Context Learning and Chain-of-Thought can improve safety, these methods still struggle to identify unintentional attacks. In addition, larger models demonstrate better safety capabilities, particularly in refusing dangerous commands. Our findings and benchmark can facilitate the design and evaluation of physical safety for LLMs. The project page is available at huggingface.co/spaces/TrustSafeAI/LLM-physical-safety.

Large Language Models (LLMs) have revolutionized various domains but remain vulnerable to prompt injection attacks, where malicious inputs manipulate the model into ignoring original instructions and executing designated action. In this paper, we investigate the underlying mechanisms of these attacks by analyzing the attention patterns within LLMs. We introduce the concept of the distraction effect, where specific attention heads, termed important heads, shift focus from the original instruction to the injected instruction. Building on this discovery, we propose Attention Tracker, a training-free detection method that tracks attention patterns on instruction to detect prompt injection attacks without the need for additional LLM inference. Our method generalizes effectively across diverse models, datasets, and attack types, showing an AUROC improvement of up to 10.0% over existing methods, and performs well even on small LLMs. We demonstrate the robustness of our approach through extensive evaluations and provide insights into safeguarding LLM-integrated systems from prompt injection vulnerabilities., Comment: Project page: https://huggingface.co/spaces/TrustSafeAI/Attention-Tracker

Laboratory accidents pose significant risks to human life and property, underscoring the importance of robust safety protocols. Despite advancements in safety training, laboratory personnel may still unknowingly engage in unsafe practices. With the increasing reliance on large language models (LLMs) for guidance in various fields, including laboratory settings, there is a growing concern about their reliability in critical safety-related decision-making. Unlike trained human researchers, LLMs lack formal lab safety education, raising questions about their ability to provide safe and accurate guidance. Existing research on LLM trustworthiness primarily focuses on issues such as ethical compliance, truthfulness, and fairness but fails to fully cover safety-critical real-world applications, like lab safety. To address this gap, we propose the Laboratory Safety Benchmark (LabSafety Bench), a comprehensive evaluation framework based on a new taxonomy aligned with Occupational Safety and Health Administration (OSHA) protocols. This benchmark includes 765 multiple-choice questions verified by human experts, assessing LLMs and vision language models (VLMs) performance in lab safety contexts. Our evaluations demonstrate that while GPT-4o outperforms human participants, it is still prone to critical errors, highlighting the risks of relying on LLMs in safety-critical environments. Our findings emphasize the need for specialized benchmarks to accurately assess the trustworthiness of LLMs in real-world safety applications., Comment: 50 pages, 19 figures

Understanding the structural and functional characteristics of proteins are crucial for developing preventative and curative strategies that impact fields from drug discovery to policy development. An important and popular technique for examining how amino acids make up these characteristics of the protein sequences with position-specific scoring (PSS). While the string kernel is crucial in natural language processing (NLP), it is unclear if string kernels can extract biologically meaningful information from protein sequences, despite the fact that they have been shown to be effective in the general sequence analysis tasks. In this work, we propose a weighted PSS kernel matrix (or W-PSSKM), that combines a PSS representation of protein sequences, which encodes the frequency information of each amino acid in a sequence, with the notion of the string kernel. This results in a novel kernel function that outperforms many other approaches for protein sequence classification. We perform extensive experimentation to evaluate the proposed method. Our findings demonstrate that the W-PSSKM significantly outperforms existing baselines and state-of-the-art methods and achieves up to 45.1\% improvement in classification accuracy.

Fine-tuning on task-specific data to boost downstream performance is a crucial step for leveraging Large Language Models (LLMs). However, previous studies have demonstrated that fine-tuning the models on several adversarial samples or even benign data can greatly comprise the model's pre-equipped alignment and safety capabilities. In this work, we propose SEAL, a novel framework to enhance safety in LLM fine-tuning. SEAL learns a data ranker based on the bilevel optimization to up rank the safe and high-quality fine-tuning data and down rank the unsafe or low-quality ones. Models trained with SEAL demonstrate superior quality over multiple baselines, with 8.5% and 9.7% win rate increase compared to random selection respectively on Llama-3-8b-Instruct and Merlinite-7b models. Our code is available on github https://github.com/hanshen95/SEAL.

Recent advances in Text-to-Speech (TTS) and Voice-Conversion (VC) using generative Artificial Intelligence (AI) technology have made it possible to generate high-quality and realistic human-like audio. This introduces significant challenges to distinguishing AI-synthesized speech from the authentic human voice and could raise potential issues of misuse for malicious purposes such as impersonation and fraud, spreading misinformation, deepfakes, and scams. However, existing detection techniques for AI-synthesized audio have not kept pace and often exhibit poor generalization across diverse datasets. In this paper, we introduce SONAR, a synthetic AI-Audio Detection Framework and Benchmark, aiming to provide a comprehensive evaluation for distinguishing cutting-edge AI-synthesized auditory content. SONAR includes a novel evaluation dataset sourced from 9 diverse audio synthesis platforms, including leading TTS providers and state-of-the-art TTS models. It is the first framework to uniformly benchmark AI-audio detection across both traditional and foundation model-based deepfake detection systems. Through extensive experiments, we reveal the generalization limitations of existing detection methods and demonstrate that foundation models exhibit stronger generalization capabilities, which can be attributed to their model size and the scale and quality of pretraining data. Additionally, we explore the effectiveness and efficiency of few-shot fine-tuning in improving generalization, highlighting its potential for tailored applications, such as personalized detection systems for specific entities or individuals. Code and dataset are available at https://github.com/Jessegator/SONAR.

Reducing the likelihood of generating harmful and toxic output is an essential task when aligning large language models (LLMs). Existing methods mainly rely on training an external reward model (i.e., another language model) or fine-tuning the LLM using self-generated data to influence the outcome. In this paper, we show that LLMs have the capability of self-detoxification without the use of an additional reward model or re-training. We propose \textit{Self-disciplined Autoregressive Sampling (SASA)}, a lightweight controlled decoding algorithm for toxicity reduction of LLMs. SASA leverages the contextual representations from an LLM to learn linear subspaces characterizing toxic v.s. non-toxic output in analytical forms. When auto-completing a response token-by-token, SASA dynamically tracks the margin of the current output to steer the generation away from the toxic subspace, by adjusting the autoregressive sampling strategy. Evaluated on LLMs of different scale and nature, namely Llama-3.1-Instruct (8B), Llama-2 (7B), and GPT2-L models with the RealToxicityPrompts, BOLD, and AttaQ benchmarks, SASA markedly enhances the quality of the generated sentences relative to the original models and attains comparable performance to state-of-the-art detoxification techniques, significantly reducing the toxicity level by only using the LLM's internal representations., Comment: 20 pages

LLM-as-a-Judge has been widely utilized as an evaluation method in various benchmarks and served as supervised rewards in model training. However, despite their excellence in many domains, potential issues are under-explored, undermining their reliability and the scope of their utility. Therefore, we identify 12 key potential biases and propose a new automated bias quantification framework-CALM-which systematically quantifies and analyzes each type of bias in LLM-as-a-Judge by using automated and principle-guided modification. Our experiments cover multiple popular language models, and the results indicate that while advanced models have achieved commendable overall performance, significant biases persist in certain specific tasks. Empirical results suggest that there remains room for improvement in the reliability of LLM-as-a-Judge. Moreover, we also discuss the explicit and implicit influence of these biases and give some suggestions for the reliable application of LLM-as-a-Judge. Our work highlights the need for stakeholders to address these issues and remind users to exercise caution in LLM-as-a-Judge applications.

Chain-of-Thought (CoT) is an efficient prompting method that enables the reasoning ability of large language models by augmenting the query using multiple examples with multiple intermediate steps. Despite the empirical success, the theoretical understanding of how to train a Transformer to achieve the CoT ability remains less explored. This is primarily due to the technical challenges involved in analyzing the nonconvex optimization on nonlinear attention models. To the best of our knowledge, this work provides the first theoretical study of training Transformers with nonlinear attention to obtain the CoT generalization capability so that the resulting model can inference on unseen tasks when the input is augmented by examples of the new task. We first quantify the required training samples and iterations to train a Transformer model towards CoT ability. We then prove the success of its CoT generalization on unseen tasks with distribution-shifted testing data. Moreover, we theoretically characterize the conditions for an accurate reasoning output by CoT even when the provided reasoning examples contain noises and are not always accurate. In contrast, in-context learning (ICL), which can be viewed as one-step CoT without intermediate steps, may fail to provide an accurate output when CoT does. These theoretical findings are justified through experiments.

As generative AI, particularly large language models (LLMs), become increasingly integrated into production applications, new attack surfaces and vulnerabilities emerge and put a focus on adversarial threats in natural language and multi-modal systems. Red-teaming has gained importance in proactively identifying weaknesses in these systems, while blue-teaming works to protect against such adversarial attacks. Despite growing academic interest in adversarial risks for generative AI, there is limited guidance tailored for practitioners to assess and mitigate these challenges in real-world environments. To address this, our contributions include: (1) a practical examination of red- and blue-teaming strategies for securing generative AI, (2) identification of key challenges and open questions in defense development and evaluation, and (3) the Attack Atlas, an intuitive framework that brings a practical approach to analyzing single-turn input attacks, placing it at the forefront for practitioners. This work aims to bridge the gap between academic insights and practical security measures for the protection of generative AI systems.

Adapting pre-trained models to new tasks can exhibit varying effectiveness across datasets. Visual prompting, a state-of-the-art parameter-efficient transfer learning method, can significantly improve the performance of out-of-distribution tasks. On the other hand, linear probing, a standard transfer learning method, can sometimes become the best approach. We propose a log-likelihood ratio (LLR) approach to analyze the comparative benefits of visual prompting and linear probing. By employing the LLR score alongside resource-efficient visual prompts approximations, our cost-effective measure attains up to a 100-fold reduction in run time compared to full training, while achieving prediction accuracies up to 91%. The source code is available at https://github.com/IBM/VP-LLR.

Most evaluations of vision models use indirect methods to assess latent space quality. These methods often involve adding extra layers to project the latent space into a new one. This projection makes it difficult to analyze and compare the original latent space. This article uses the k* Distribution, a local neighborhood analysis method, to examine the learned latent space at the level of individual concepts, which can be extended to examine the entire latent space. We introduce skewness-based true and approximate metrics for interpreting individual concepts to assess the overall quality of vision models' latent space. Our findings indicate that current vision models frequently fracture the distributions of individual concepts within the latent space. Nevertheless, as these models improve in generalization across multiple datasets, the degree of fracturing diminishes. A similar trend is observed in robust vision models, where increased robustness correlates with reduced fracturing. Ultimately, this approach enables a direct interpretation and comparison of the latent spaces of different vision models and reveals a relationship between a model's generalizability and robustness. Results show that as a model becomes more general and robust, it tends to learn features that result in better clustering of concepts. Project Website is available online at https://shashankkotyan.github.io/k-Distribution/

Sepsis is the leading cause of in-hospital mortality in the USA. Early sepsis onset prediction and diagnosis could significantly improve the survival of sepsis patients. Existing predictive models are usually trained on high-quality data with few missing information, while missing values widely exist in real-world clinical scenarios (especially in the first hours of admissions to the hospital), which causes a significant decrease in accuracy and an increase in uncertainty for the predictive models. The common method to handle missing values is imputation, which replaces the unavailable variables with estimates from the observed data. The uncertainty of imputation results can be propagated to the sepsis prediction outputs, which have not been studied in existing works on either sepsis prediction or uncertainty quantification. In this study, we first define such propagated uncertainty as the variance of prediction output and then introduce uncertainty propagation methods to quantify the propagated uncertainty. Moreover, for the potential high-risk patients with low confidence due to limited observations, we propose a robust active sensing algorithm to increase confidence by actively recommending clinicians to observe the most informative variables. We validate the proposed models in both publicly available data (i.e., MIMIC-III and AmsterdamUMCdb) and proprietary data in The Ohio State University Wexner Medical Center (OSUWMC). The experimental results show that the propagated uncertainty is dominant at the beginning of admissions to hospitals and the proposed algorithm outperforms state-of-the-art active sensing methods. Finally, we implement a SepsisLab system for early sepsis prediction and active sensing based on our pre-trained models. Clinicians and potential sepsis patients can benefit from the system in early prediction and diagnosis of sepsis., Comment: To be published in KDD 2024

The security and robustness of deep neural networks (DNNs) have become increasingly concerning. This paper aims to provide both a theoretical foundation and a practical solution to ensure the reliability of DNNs. We explore the concept of Lipschitz continuity to certify the robustness of DNNs against adversarial attacks, which aim to mislead the network with adding imperceptible perturbations into inputs. We propose a novel algorithm that remaps the input domain into a constrained range, reducing the Lipschitz constant and potentially enhancing robustness. Unlike existing adversarially trained models, where robustness is enhanced by introducing additional examples from other datasets or generative models, our method is almost cost-free as it can be integrated with existing models without requiring re-training. Experimental results demonstrate the generalizability of our method, as it can be combined with various models and achieve enhancements in robustness. Furthermore, our method achieves the best robust accuracy for CIFAR10, CIFAR100, and ImageNet datasets on the RobustBench leaderboard.

Efficient training and inference algorithms, such as low-rank adaption and model pruning, have shown impressive performance for learning Transformer-based large foundation models. However, due to the technical challenges of the non-convex optimization caused by the complicated architecture of Transformers, the theoretical study of why these methods can be applied to learn Transformers is mostly elusive. To the best of our knowledge, this paper shows the first theoretical analysis of the property of low-rank and sparsity of one-layer Transformers by characterizing the trained model after convergence using stochastic gradient descent. By focusing on a data model based on label-relevant and label-irrelevant patterns, we quantify that the gradient updates of trainable parameters are low-rank, which depends on the number of label-relevant patterns. We also analyze how model pruning affects the generalization while improving computation efficiency and conclude that proper magnitude-based pruning has a slight effect on the testing performance. We implement numerical experiments to support our findings., Comment: IEEE SAM Workshop 2024

Pre-trained Language models (PLMs) have been acknowledged to contain harmful information, such as social biases, which may cause negative social impacts or even bring catastrophic results in application. Previous works on this problem mainly focused on using black-box methods such as probing to detect and quantify social biases in PLMs by observing model outputs. As a result, previous debiasing methods mainly finetune or even pre-train language models on newly constructed anti-stereotypical datasets, which are high-cost. In this work, we try to unveil the mystery of social bias inside language models by introducing the concept of {\sc Social Bias Neurons}. Specifically, we propose {\sc Integrated Gap Gradients (IG$^2$)} to accurately pinpoint units (i.e., neurons) in a language model that can be attributed to undesirable behavior, such as social bias. By formalizing undesirable behavior as a distributional property of language, we employ sentiment-bearing prompts to elicit classes of sensitive words (demographics) correlated with such sentiments. Our IG$^2$ thus attributes the uneven distribution for different demographics to specific Social Bias Neurons, which track the trail of unwanted behavior inside PLM units to achieve interoperability. Moreover, derived from our interpretable technique, {\sc Bias Neuron Suppression (BNS)} is further proposed to mitigate social biases. By studying BERT, RoBERTa, and their attributable differences from debiased FairBERTa, IG$^2$ allows us to locate and suppress identified neurons, and further mitigate undesired behaviors. As measured by prior metrics from StereoSet, our model achieves a higher degree of fairness while maintaining language modeling ability with low cost.

Deep neural networks are susceptible to backdoor attacks, where adversaries manipulate model predictions by inserting malicious samples into the training data. Currently, there is still a lack of direct filtering methods for identifying suspicious training data to unveil potential backdoor samples. In this paper, we propose a novel method, Prediction Shift Backdoor Detection (PSBD), leveraging an uncertainty-based approach requiring minimal unlabeled clean validation data. PSBD is motivated by an intriguing Prediction Shift (PS) phenomenon, where poisoned models' predictions on clean data often shift away from true labels towards certain other labels with dropout applied during inference, while backdoor samples exhibit less PS. We hypothesize PS results from neuron bias effect, making neurons favor features of certain classes. PSBD identifies backdoor training samples by computing the Prediction Shift Uncertainty (PSU), the variance in probability values when dropout layers are toggled on and off during model inference. Extensive experiments have been conducted to verify the effectiveness and efficiency of PSBD, which achieves state-of-the-art results among mainstream detection methods.

Large language models are first pre-trained on trillions of tokens and then instruction-tuned or aligned to specific preferences. While pre-training remains out of reach for most researchers due to the compute required, fine-tuning has become affordable thanks to parameter-efficient methods such as LoRA and QLoRA. Alignment is known to be sensitive to the many factors involved, including the quantity and quality of data, the alignment method, and the adapter rank. However, there has not yet been an extensive study of their effect on downstream performance. To address this gap, we conduct an in-depth investigation of the impact of popular choices for three crucial axes: (i) the alignment dataset (HH-RLHF and BeaverTails), (ii) the alignment technique (SFT and DPO), and (iii) the model (LLaMA-1, Vicuna-v1.3, Mistral-7b, and Mistral-7b-Instruct). Our extensive setup spanning over 300 experiments reveals consistent trends and unexpected findings. We observe how more informative data helps with preference alignment, cases where supervised fine-tuning outperforms preference optimization, and how aligning to a distinct preference boosts performance on downstream tasks. Through our in-depth analyses, we put forward key guidelines to help researchers perform more effective parameter-efficient LLM alignment., Comment: Accepted to ACL (Main) 2024

Jailbreak attacks in large language models (LLMs) entail inducing the models to generate content that breaches ethical and legal norm through the use of malicious prompts, posing a substantial threat to LLM security. Current strategies for jailbreak attack and defense often focus on optimizing locally within specific algorithmic frameworks, resulting in ineffective optimization and limited scalability. In this paper, we present a systematic analysis of the dependency relationships in jailbreak attack and defense techniques, generalizing them to all possible attack surfaces. We employ directed acyclic graphs (DAGs) to position and analyze existing jailbreak attacks, defenses, and evaluation methodologies, and propose three comprehensive, automated, and logical frameworks. \texttt{AutoAttack} investigates dependencies in two lines of jailbreak optimization strategies: genetic algorithm (GA)-based attacks and adversarial-generation-based attacks, respectively. We then introduce an ensemble jailbreak attack to exploit these dependencies. \texttt{AutoDefense} offers a mixture-of-defenders approach by leveraging the dependency relationships in pre-generative and post-generative defense strategies. \texttt{AutoEvaluation} introduces a novel evaluation method that distinguishes hallucinations, which are often overlooked, from jailbreak attack and defense responses. Through extensive experiments, we demonstrate that the proposed ensemble jailbreak attack and defense framework significantly outperforms existing research., Comment: 32 pages, 2 figures

Graph Transformers, which incorporate self-attention and positional encoding, have recently emerged as a powerful architecture for various graph learning tasks. Despite their impressive performance, the complex non-convex interactions across layers and the recursive graph structure have made it challenging to establish a theoretical foundation for learning and generalization. This study introduces the first theoretical investigation of a shallow Graph Transformer for semi-supervised node classification, comprising a self-attention layer with relative positional encoding and a two-layer perceptron. Focusing on a graph data model with discriminative nodes that determine node labels and non-discriminative nodes that are class-irrelevant, we characterize the sample complexity required to achieve a desirable generalization error by training with stochastic gradient descent (SGD). This paper provides the quantitative characterization of the sample complexity and number of iterations for convergence dependent on the fraction of discriminative nodes, the dominant patterns, and the initial model errors. Furthermore, we demonstrate that self-attention and positional encoding enhance generalization by making the attention map sparse and promoting the core neighborhood during training, which explains the superior feature representation of Graph Transformers. Our theoretical results are supported by empirical experiments on synthetic and real-world benchmarks., Comment: ICML 2024

WiFi-based indoor positioning has been extensively studied. A fundamental issue in such solutions is the collection of WiFi fingerprints. However, due to real-world constraints, collecting complete fingerprints at all intended locations is sometimes prohibited. This work considers the WiFi fingerprint inpainting problem. This problem differs from typical image/video inpainting problems in several aspects. Unlike RGB images, WiFi field maps come in any shape, and signal data may follow certain distributions. Therefore, it is difficult to forcefully fit them into a fixed-dimensional matrix, as done with processing images in RGB format. As soon as a map is changed, it also becomes difficult to adapt it to the same model due to scale issues. Furthermore, such models are significantly constrained in situations requiring outward inpainting. Fortunately, the spatial relationships of WiFi signals and the rich information provided among channels offer ample opportunities for this generative model to accomplish inpainting. Therefore, we designed this model to not only retain the characteristic of regression models in generating fingerprints of arbitrary shapes but also to accommodate the observational outcomes from densely deployed APs. This work makes two major contributions. Firstly, we delineate the distinctions between this problem and image inpainting, highlighting potential avenues for research. Secondly, we introduce novel generative inpainting models aimed at capturing both inter-AP and intra-AP correlations while preserving latent information. Additionally, we incorporate a specially designed adversarial discriminator to enhance the quality of inpainting outcomes., Comment: ICCCN2024

The rapid advances in generative AI models have empowered the creation of highly realistic images with arbitrary content, raising concerns about potential misuse and harm, such as Deepfakes. Current research focuses on training detectors using large datasets of generated images. However, these training-based solutions are often computationally expensive and show limited generalization to unseen generated images. In this paper, we propose a training-free method to distinguish between real and AI-generated images. We first observe that real images are more robust to tiny noise perturbations than AI-generated images in the representation space of vision foundation models. Based on this observation, we propose RIGID, a training-free and model-agnostic method for robust AI-generated image detection. RIGID is a simple yet effective approach that identifies whether an image is AI-generated by comparing the representation similarity between the original and the noise-perturbed counterpart. Our evaluation on a diverse set of AI-generated images and benchmarks shows that RIGID significantly outperforms existing trainingbased and training-free detectors. In particular, the average performance of RIGID exceeds the current best training-free method by more than 25%. Importantly, RIGID exhibits strong generalization across different image generation methods and robustness to image corruptions.

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.

The exposure of security vulnerabilities in safety-aligned language models, e.g., susceptibility to adversarial attacks, has shed light on the intricate interplay between AI safety and AI security. Although the two disciplines now come together under the overarching goal of AI risk management, they have historically evolved separately, giving rise to differing perspectives. Therefore, in this paper, we advocate that stakeholders in AI risk management should be aware of the nuances, synergies, and interplay between safety and security, and unambiguously take into account the perspectives of both disciplines in order to devise mostly effective and holistic risk mitigation approaches. Unfortunately, this vision is often obfuscated, as the definitions of the basic concepts of "safety" and "security" themselves are often inconsistent and lack consensus across communities. With AI risk management being increasingly cross-disciplinary, this issue is particularly salient. In light of this conceptual challenge, we introduce a unified reference framework to clarify the differences and interplay between AI safety and AI security, aiming to facilitate a shared understanding and effective collaboration across communities.

Safety alignment is crucial to ensure that large language models (LLMs) behave in ways that align with human preferences and prevent harmful actions during inference. However, recent studies show that the alignment can be easily compromised through finetuning with only a few adversarially designed training examples. We aim to measure the risks in finetuning LLMs through navigating the LLM safety landscape. We discover a new phenomenon observed universally in the model parameter space of popular open-source LLMs, termed as "safety basin": random perturbations to model weights maintain the safety level of the original aligned model within its local neighborhood. However, outside this local region, safety is fully compromised, exhibiting a sharp, step-like drop. This safety basin contrasts sharply with the LLM capability landscape, where model performance peaks at the origin and gradually declines as random perturbation increases. Our discovery inspires us to propose the new VISAGE safety metric that measures the safety in LLM finetuning by probing its safety landscape. Visualizing the safety landscape of the aligned model enables us to understand how finetuning compromises safety by dragging the model away from the safety basin. The LLM safety landscape also highlights the system prompt's critical role in protecting a model, and that such protection transfers to its perturbed variants within the safety basin. These observations from our safety landscape research provide new insights for future work on LLM safety community. Our code is publicly available at https://github.com/ShengYun-Peng/llm-landscape., Comment: NeurIPS'24

While large language models (LLMs) such as Llama-2 or GPT-4 have shown impressive zero-shot performance, fine-tuning is still necessary to enhance their performance for customized datasets, domain-specific tasks, or other private needs. However, fine-tuning all parameters of LLMs requires significant hardware resources, which can be impractical for typical users. Therefore, parameter-efficient fine-tuning such as LoRA have emerged, allowing users to fine-tune LLMs without the need for considerable computing resources, with little performance degradation compared to fine-tuning all parameters. Unfortunately, recent studies indicate that fine-tuning can increase the risk to the safety of LLMs, even when data does not contain malicious content. To address this challenge, we propose Safe LoRA, a simple one-liner patch to the original LoRA implementation by introducing the projection of LoRA weights from selected layers to the safety-aligned subspace, effectively reducing the safety risks in LLM fine-tuning while maintaining utility. It is worth noting that Safe LoRA is a training-free and data-free approach, as it only requires the knowledge of the weights from the base and aligned LLMs. Our extensive experiments demonstrate that when fine-tuning on purely malicious data, Safe LoRA retains similar safety performance as the original aligned model. Moreover, when the fine-tuning dataset contains a mixture of both benign and malicious data, Safe LoRA mitigates the negative effect made by malicious data while preserving performance on downstream tasks.

The sparsely gated mixture of experts (MoE) architecture sends different inputs to different subnetworks, i.e., experts, through trainable routers. MoE reduces the training computation significantly for large models, but its deployment can be still memory or computation expensive for some downstream tasks. Model pruning is a popular approach to reduce inference computation, but its application in MoE architecture is largely unexplored. To the best of our knowledge, this paper provides the first provably efficient technique for pruning experts in finetuned MoE models. We theoretically prove that prioritizing the pruning of the experts with a smaller change of the routers l2 norm from the pretrained model guarantees the preservation of test accuracy, while significantly reducing the model size and the computational requirements. Although our theoretical analysis is centered on binary classification tasks on simplified MoE architecture, our expert pruning method is verified on large vision MoE models such as VMoE and E3MoE finetuned on benchmark datasets such as CIFAR10, CIFAR100, and ImageNet.

This paper studies the transfer reinforcement learning (RL) problem where multiple RL problems have different reward functions but share the same underlying transition dynamics. In this setting, the Q-function of each RL problem (task) can be decomposed into a successor feature (SF) and a reward mapping: the former characterizes the transition dynamics, and the latter characterizes the task-specific reward function. This Q-function decomposition, coupled with a policy improvement operator known as generalized policy improvement (GPI), reduces the sample complexity of finding the optimal Q-function, and thus the SF \& GPI framework exhibits promising empirical performance compared to traditional RL methods like Q-learning. However, its theoretical foundations remain largely unestablished, especially when learning the successor features using deep neural networks (SF-DQN). This paper studies the provable knowledge transfer using SFs-DQN in transfer RL problems. We establish the first convergence analysis with provable generalization guarantees for SF-DQN with GPI. The theory reveals that SF-DQN with GPI outperforms conventional RL approaches, such as deep Q-network, in terms of both faster convergence rate and better generalization. Numerical experiments on real and synthetic RL tasks support the superior performance of SF-DQN \& GPI, aligning with our theoretical findings., Comment: arXiv admin note: text overlap with arXiv:2310.16173

We propose an unsupervised adaptation framework, Self-TAught Recognizer (STAR), which leverages unlabeled data to enhance the robustness of automatic speech recognition (ASR) systems in diverse target domains, such as noise and accents. STAR is developed for prevalent speech foundation models based on Transformer-related architecture with auto-regressive decoding (e.g., Whisper, Canary). Specifically, we propose a novel indicator that empirically integrates step-wise information during decoding to assess the token-level quality of pseudo labels without ground truth, thereby guiding model updates for effective unsupervised adaptation. Experimental results show that STAR achieves an average of 13.5% relative reduction in word error rate across 14 target domains, and it sometimes even approaches the upper-bound performance of supervised adaptation. Surprisingly, we also observe that STAR prevents the adapted model from the common catastrophic forgetting problem without recalling source-domain data. Furthermore, STAR exhibits high data efficiency that only requires less than one-hour unlabeled data, and seamless generality to alternative large speech models and speech translation tasks. Our code aims to open source to the research communities., Comment: 23 pages, Preprint

We propose a new positional encoding method for a neural network architecture called the Transformer. Unlike the standard sinusoidal positional encoding, our approach is based on solid mathematical grounds and has a guarantee of not losing information about the positional order of the input sequence. We show that the new encoding approach systematically improves the prediction performance in the time-series classification task., Comment: arXiv admin note: text overlap with arXiv:2305.17149

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.

Latency attacks against object detection represent a variant of adversarial attacks that aim to inflate the inference time by generating additional ghost objects in a target image. However, generating ghost objects in the black-box scenario remains a challenge since information about these unqualified objects remains opaque. In this study, we demonstrate the feasibility of generating ghost objects in adversarial examples by extending the concept of "steal now, decrypt later" attacks. These adversarial examples, once produced, can be employed to exploit potential vulnerabilities in the AI service, giving rise to significant security concerns. The experimental results demonstrate that the proposed attack achieves successful attacks across various commonly used models and Google Vision API without any prior knowledge about the target model. Additionally, the average cost of each attack is less than \$ 1 dollars, posing a significant threat to AI security.