A Survey on Data Synthesis and Augmentation for Large Language Models

The success of Large Language Models (LLMs) is inherently linked to the availability of vast, diverse, and high-quality data for training and evaluation. However, the growth rate of high-quality data is significantly outpaced by the expansion of training datasets, leading to a looming data exhaustion crisis. This underscores the urgent need to enhance data efficiency and explore new data sources. In this context, synthetic data has emerged as a promising solution. Currently, data generation primarily consists of two major approaches: data augmentation and synthesis. This paper comprehensively reviews and summarizes data generation techniques throughout the lifecycle of LLMs, including data preparation, pre-training, fine-tuning, instruction-tuning, preference alignment, and applications. Furthermore, We discuss the current constraints faced by these methods and investigate potential pathways for future development and research. Our aspiration is to equip researchers with a clear understanding of these methodologies, enabling them to swiftly identify appropriate data generation strategies in the construction of LLMs, while providing valuable insights for future exploration.

Revealing the Unseen: Guiding Personalized Diffusion Models to Expose Training Data

Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot fine-tuning where a pretrained DM is fine-tuned on a small set of images to capture specific styles or objects. Many people upload these personalized checkpoints online, fostering communities such as Civitai and HuggingFace. However, model owners may overlook the potential risks of data leakage by releasing their fine-tuned checkpoints. Moreover, concerns regarding copyright violations arise when unauthorized data is used during fine-tuning. In this paper, we ask: "Can training data be extracted from these fine-tuned DMs shared online?" A successful extraction would present not only data leakage threats but also offer tangible evidence of copyright infringement. To answer this, we propose FineXtract, a framework for extracting fine-tuning data. Our method approximates fine-tuning as a gradual shift in the model's learned distribution -- from the original pretrained DM toward the fine-tuning data. By extrapolating the models before and after fine-tuning, we guide the generation toward high-probability regions within the fine-tuned data distribution. We then apply a clustering algorithm to extract the most probable images from those generated using this extrapolated guidance. Experiments on DMs fine-tuned with datasets such as WikiArt, DreamBooth, and real-world checkpoints posted online validate the effectiveness of our method, extracting approximately 20% of fine-tuning data in most cases, significantly surpassing baseline performance.

Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks

Few-shot fine-tuning of Diffusion Models (DMs) is a key advancement, significantly reducing training costs and enabling personalized AI applications. However, we explore the training dynamics of DMs and observe an unanticipated phenomenon: during the training process, image fidelity initially improves, then unexpectedly deteriorates with the emergence of noisy patterns, only to recover later with severe overfitting. We term the stage with generated noisy patterns as corruption stage. To understand this corruption stage, we begin by theoretically modeling the one-shot fine-tuning scenario, and then extend this modeling to more general cases. Through this modeling, we identify the primary cause of this corruption stage: a narrowed learning distribution inherent in the nature of few-shot fine-tuning. To tackle this, we apply Bayesian Neural Networks (BNNs) on DMs with variational inference to implicitly broaden the learned distribution, and present that the learning target of the BNNs can be naturally regarded as an expectation of the diffusion loss and a further regularization with the pretrained DMs. This approach is highly compatible with current few-shot fine-tuning methods in DMs and does not introduce any extra inference costs. Experimental results demonstrate that our method significantly mitigates corruption, and improves the fidelity, quality and diversity of the generated images in both object-driven and subject-driven generation tasks.

CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion

Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot generation where a pretrained model is fine-tuned on a small set of images to capture a specific style or object. Despite their success, concerns exist about potential copyright violations stemming from the use of unauthorized data in this process. In response, we present Contrasting Gradient Inversion for Diffusion Models (CGI-DM), a novel method featuring vivid visual representations for digital copyright authentication. Our approach involves removing partial information of an image and recovering missing details by exploiting conceptual differences between the pretrained and fine-tuned models. We formulate the differences as KL divergence between latent variables of the two models when given the same input image, which can be maximized through Monte Carlo sampling and Projected Gradient Descent (PGD). The similarity between original and recovered images serves as a strong indicator of potential infringements. Extensive experiments on the WikiArt and Dreambooth datasets demonstrate the high accuracy of CGI-DM in digital copyright authentication, surpassing alternative validation techniques. Code implementation is available at https://github.com/Nicholas0228/Revelio.

Understanding and Improving Adversarial Attacks on Latent Diffusion Model

Latent Diffusion Model (LDM) has emerged as a leading tool in image generation, particularly with its capability in few-shot generation. This capability also presents risks, notably in unauthorized artwork replication and misinformation generation. In response, adversarial attacks have been designed to safeguard personal images from being used as reference data. However, existing adversarial attacks are predominantly empirical, lacking a solid theoretical foundation. In this paper, we introduce a comprehensive theoretical framework for understanding adversarial attacks on LDM. Based on the framework, we propose a novel adversarial attack that exploits a unified target to guide the adversarial attack both in the forward and the reverse process of LDM. We provide empirical evidences that our method overcomes the offset problem of the optimization of adversarial attacks in existing methods. Through rigorous experiments, our findings demonstrate that our method outperforms current attacks and is able to generalize over different state-of-the-art few-shot generation pipelines based on LDM. Our method can serve as a stronger and efficient tool for people exposed to the risk of data privacy and security to protect themselves in the new era of powerful generative models. The code is available on GitHub: https://github.com/CaradryanLiang/ImprovedAdvDM.git.

What Makes Good Open-Vocabulary Detector: A Disassembling Perspective

Open-vocabulary detection (OVD) is a new object detection paradigm, aiming to localize and recognize unseen objects defined by an unbounded vocabulary. This is challenging since traditional detectors can only learn from pre-defined categories and thus fail to detect and localize objects out of pre-defined vocabulary. To handle the challenge, OVD leverages pre-trained cross-modal VLM, such as CLIP, ALIGN, etc. Previous works mainly focus on the open vocabulary classification part, with less attention on the localization part. We argue that for a good OVD detector, both classification and localization should be parallelly studied for the novel object categories. We show in this work that improving localization as well as cross-modal classification complement each other, and compose a good OVD detector jointly. We analyze three families of OVD methods with different design emphases. We first propose a vanilla method,i.e., cropping a bounding box obtained by a localizer and resizing it into the CLIP. We next introduce another approach, which combines a standard two-stage object detector with CLIP. A two-stage object detector includes a visual backbone, a region proposal network (RPN), and a region of interest (RoI) head. We decouple RPN and ROI head (DRR) and use RoIAlign to extract meaningful features. In this case, it avoids resizing objects. To further accelerate the training time and reduce the model parameters, we couple RPN and ROI head (CRR) as the third approach. We conduct extensive experiments on these three types of approaches in different settings. On the OVD-COCO benchmark, DRR obtains the best performance and achieves 35.8 Novel AP$_{50}$, an absolute 2.8 gain over the previous state-of-the-art (SOTA). For OVD-LVIS, DRR surpasses the previous SOTA by 1.9 AP$_{50}$ in rare categories. We also provide an object detection dataset called PID and provide a baseline on PID.

Learning Prompt-Enhanced Context Features for Weakly-Supervised Video Anomaly Detection

Video anomaly detection under weak supervision is challenging due to the absence of frame-level annotations during the training phase. Previous work has employed graph convolution networks or self-attention mechanisms to model temporal relations, along with multiple instance learning (MIL)-based classification loss to learn discriminative features. However, most of them utilize multi-branches to capture local and global dependencies separately, leading to increased parameters and computational cost. Furthermore, the binarized constraint of the MIL-based loss only ensures coarse-grained interclass separability, ignoring fine-grained discriminability within anomalous classes. In this paper, we propose a weakly supervised anomaly detection framework that emphasizes efficient context modeling and enhanced semantic discriminability. To this end, we first construct a temporal context aggregation (TCA) module that captures complete contextual information by reusing similarity matrix and adaptive fusion. Additionally, we propose a prompt-enhanced learning (PEL) module that incorporates semantic priors into the model by utilizing knowledge-based prompts, aiming at enhancing the discriminative capacity of context features while ensuring separability between anomaly sub-classes. Furthermore, we introduce a score smoothing (SS) module in the testing phase to suppress individual bias and reduce false alarms. Extensive experiments demonstrate the effectiveness of various components of our method, which achieves competitive performance with fewer parameters and computational effort on three challenging benchmarks: the UCF-crime, XD-violence, and ShanghaiTech datasets. The detection accuracy of some anomaly sub-classes is also improved with a great margin.

Mist: Towards Improved Adversarial Examples for Diffusion Models

Diffusion Models (DMs) have empowered great success in artificial-intelligence-generated content, especially in artwork creation, yet raising new concerns in intellectual properties and copyright. For example, infringers can make profits by imitating non-authorized human-created paintings with DMs. Recent researches suggest that various adversarial examples for diffusion models can be effective tools against these copyright infringements. However, current adversarial examples show weakness in transferability over different painting-imitating methods and robustness under straightforward adversarial defense, for example, noise purification. We surprisingly find that the transferability of adversarial examples can be significantly enhanced by exploiting a fused and modified adversarial loss term under consistent parameters. In this work, we comprehensively evaluate the cross-method transferability of adversarial examples. The experimental observation shows that our method generates more transferable adversarial examples with even stronger robustness against the simple adversarial defense.

Adversarial Example Does Good: Preventing Painting Imitation from Diffusion Models via Adversarial Examples

Diffusion Models (DMs) achieve state-of-the-art performance in generative tasks, boosting a wave in AI for Art. Despite the success of commercialization, DMs meanwhile provide tools for copyright violations, where infringers benefit from illegally using paintings created by human artists to train DMs and generate novel paintings in a similar style. In this paper, we show that it is possible to create an image $x'$ that is similar to an image $x$ for human vision but unrecognizable for DMs. We build a framework to define and evaluate this adversarial example for diffusion models. Based on the framework, we further propose AdvDM, an algorithm to generate adversarial examples for DMs. By optimizing upon different latent variables sampled from the reverse process of DMs, AdvDM conducts a Monte-Carlo estimation of adversarial examples for DMs. Extensive experiments show that the estimated adversarial examples can effectively hinder DMs from extracting their features. Our method can be a powerful tool for human artists to protect their copyright against infringers with DM-based AI-for-Art applications.

Temporal and Contextual Transformer for Multi-Camera Editing of TV Shows

The ability to choose an appropriate camera view among multiple cameras plays a vital role in TV shows delivery. But it is hard to figure out the statistical pattern and apply intelligent processing due to the lack of high-quality training data. To solve this issue, we first collect a novel benchmark on this setting with four diverse scenarios including concerts, sports games, gala shows, and contests, where each scenario contains 6 synchronized tracks recorded by different cameras. It contains 88-hour raw videos that contribute to the 14-hour edited videos. Based on this benchmark, we further propose a new approach temporal and contextual transformer that utilizes clues from historical shots and other views to make shot transition decisions and predict which view to be used. Extensive experiments show that our method outperforms existing methods on the proposed multi-camera editing benchmark.