Hollowed Net for On-Device Personalization of Text-to-Image Diffusion Models

Recent advancements in text-to-image diffusion models have enabled the personalization of these models to generate custom images from textual prompts. This paper presents an efficient LoRA-based personalization approach for on-device subject-driven generation, where pre-trained diffusion models are fine-tuned with user-specific data on resource-constrained devices. Our method, termed Hollowed Net, enhances memory efficiency during fine-tuning by modifying the architecture of a diffusion U-Net to temporarily remove a fraction of its deep layers, creating a hollowed structure. This approach directly addresses on-device memory constraints and substantially reduces GPU memory requirements for training, in contrast to previous methods that primarily focus on minimizing training steps and reducing the number of parameters to update. Additionally, the personalized Hollowed Net can be transferred back into the original U-Net, enabling inference without additional memory overhead. Quantitative and qualitative analyses demonstrate that our approach not only reduces training memory to levels as low as those required for inference but also maintains or improves personalization performance compared to existing methods.

Model-based Preference Optimization in Abstractive Summarization without Human Feedback

In abstractive summarization, the challenge of producing concise and accurate summaries arises from the vast amount of information contained in the source document. Consequently, although Large Language Models (LLMs) can generate fluent text, they often introduce inaccuracies by hallucinating content not found in the original source. While supervised fine-tuning methods that maximize likelihood contribute to this issue, they do not consistently enhance the faithfulness of the summaries. Preference-based optimization methods, such as Direct Preference Optimization (DPO), can further refine the model to align with human preferences. However, these methods still heavily depend on costly human feedback. In this work, we introduce a novel and straightforward approach called Model-based Preference Optimization (MPO) to fine-tune LLMs for improved summarization abilities without any human feedback. By leveraging the model's inherent summarization capabilities, we create a preference dataset that is fully generated by the model using different decoding strategies. Our experiments on standard summarization datasets and various metrics demonstrate that our proposed MPO significantly enhances the quality of generated summaries without relying on human feedback.

Missing Modality Prediction for Unpaired Multimodal Learning via Joint Embedding of Unimodal Models

Multimodal learning typically relies on the assumption that all modalities are fully available during both the training and inference phases. However, in real-world scenarios, consistently acquiring complete multimodal data presents significant challenges due to various factors. This often leads to the issue of missing modalities, where data for certain modalities are absent, posing considerable obstacles not only for the availability of multimodal pretrained models but also for their fine-tuning and the preservation of robustness in downstream tasks. To address these challenges, we propose a novel framework integrating parameter-efficient fine-tuning of unimodal pretrained models with a self-supervised joint-embedding learning method. This framework enables the model to predict the embedding of a missing modality in the representation space during inference. Our method effectively predicts the missing embedding through prompt tuning, leveraging information from available modalities. We evaluate our approach on several multimodal benchmark datasets and demonstrate its effectiveness and robustness across various scenarios of missing modalities.

Mitigating Hallucination in Abstractive Summarization with Domain-Conditional Mutual Information

A primary challenge in abstractive summarization is hallucination -- the phenomenon where a model generates plausible text that is absent in the source text. We hypothesize that the domain (or topic) of the source text triggers the model to generate text that is highly probable in the domain, neglecting the details of the source text. To alleviate this model bias, we introduce a decoding strategy based on domain-conditional pointwise mutual information. This strategy adjusts the generation probability of each token by comparing it with the token's marginal probability within the domain of the source text. According to evaluation on the XSUM dataset, our method demonstrates improvement in terms of faithfulness and source relevance. The code is publicly available at \url{https://github.com/qqplot/dcpmi}.

Overcoming Data Inequality across Domains with Semi-Supervised Domain Generalization

While there have been considerable advancements in machine learning driven by extensive datasets, a significant disparity still persists in the availability of data across various sources and populations. This inequality across domains poses challenges in modeling for those with limited data, which can lead to profound practical and ethical concerns. In this paper, we address a representative case of data inequality problem across domains termed Semi-Supervised Domain Generalization (SSDG), in which only one domain is labeled while the rest are unlabeled. We propose a novel algorithm, ProUD, which can effectively learn domain-invariant features via domain-aware prototypes along with progressive generalization via uncertainty-adaptive mixing of labeled and unlabeled domains. Our experiments on three different benchmark datasets demonstrate the effectiveness of ProUD, outperforming all baseline models including single domain generalization and semi-supervised learning. Source code will be released upon acceptance of the paper.

X-PEFT: eXtremely Parameter-Efficient Fine-Tuning for Extreme Multi-Profile Scenarios

Parameter-efficient fine-tuning (PEFT) techniques, such as adapter tuning, aim to fine-tune a pre-trained language model (PLM) using a minimal number of parameters for a specific task or profile. Although adapter tuning provides increased parameter efficiency compared to full-model fine-tuning, it introduces a small set of additional parameters attached to a PLM for each profile. This can become problematic in practical applications with multiple profiles, particularly when a significant increase in the number of profiles linearly boosts the total number of additional parameters. To mitigate this issue, we introduce X-PEFT, a novel PEFT method that leverages a multitude of given adapters by fine-tuning an extremely small set of compact tensors for a new profile, which serve as binary masks to adaptively select the given adapters. To efficiently validate our proposed method, we implement it using a large number of trained or untrained (random) adapters. We evaluate the performance of X-PEFT through LaMP and GLUE tasks and demonstrate that it either matches or surpasses the effectiveness of conventional adapter tuning, despite reducing the memory requirements per profile by a factor of 10,000 compared to it.

Meta-Learning with a Geometry-Adaptive Preconditioner

Model-agnostic meta-learning (MAML) is one of the most successful meta-learning algorithms. It has a bi-level optimization structure where the outer-loop process learns a shared initialization and the inner-loop process optimizes task-specific weights. Although MAML relies on the standard gradient descent in the inner-loop, recent studies have shown that controlling the inner-loop's gradient descent with a meta-learned preconditioner can be beneficial. Existing preconditioners, however, cannot simultaneously adapt in a task-specific and path-dependent way. Additionally, they do not satisfy the Riemannian metric condition, which can enable the steepest descent learning with preconditioned gradient. In this study, we propose Geometry-Adaptive Preconditioned gradient descent (GAP) that can overcome the limitations in MAML; GAP can efficiently meta-learn a preconditioner that is dependent on task-specific parameters, and its preconditioner can be shown to be a Riemannian metric. Thanks to the two properties, the geometry-adaptive preconditioner is effective for improving the inner-loop optimization. Experiment results show that GAP outperforms the state-of-the-art MAML family and preconditioned gradient descent-MAML (PGD-MAML) family in a variety of few-shot learning tasks. Code is available at: https://github.com/Suhyun777/CVPR23-GAP.

Complementary Domain Adaptation and Generalization for Unsupervised Continual Domain Shift Learning

Continual domain shift poses a significant challenge in real-world applications, particularly in situations where labeled data is not available for new domains. The challenge of acquiring knowledge in this problem setting is referred to as unsupervised continual domain shift learning. Existing methods for domain adaptation and generalization have limitations in addressing this issue, as they focus either on adapting to a specific domain or generalizing to unseen domains, but not both. In this paper, we propose Complementary Domain Adaptation and Generalization (CoDAG), a simple yet effective learning framework that combines domain adaptation and generalization in a complementary manner to achieve three major goals of unsupervised continual domain shift learning: adapting to a current domain, generalizing to unseen domains, and preventing forgetting of previously seen domains. Our approach is model-agnostic, meaning that it is compatible with any existing domain adaptation and generalization algorithms. We evaluate CoDAG on several benchmark datasets and demonstrate that our model outperforms state-of-the-art models in all datasets and evaluation metrics, highlighting its effectiveness and robustness in handling unsupervised continual domain shift learning.

Meta-Learning with Adaptive Weighted Loss for Imbalanced Cold-Start Recommendation

Sequential recommenders have made great strides in capturing a user's preferences. Nevertheless, the cold-start recommendation remains a fundamental challenge in which only a few user-item interactions are available for personalization. Gradient-based meta-learning approaches have recently emerged in the sequential recommendation field due to their fast adaptation and easy-to-integrate abilities. The meta-learning algorithms formulate the cold-start recommendation as a few-shot learning problem, where each user is represented as a task to be adapted. However, while meta-learning algorithms generally assume that task-wise samples are evenly distributed over classes or values, user-item interactions are not that way in real-world applications (e.g., watching favorite videos multiple times, leaving only good ratings and no bad ones). As a result, in the real-world, imbalanced user feedback that accounts for most task training data may dominate the user adaptation and prevent meta-learning algorithms from learning meaningful meta-knowledge for personalized recommendations. To alleviate this limitation, we propose a novel sequential recommendation framework based on gradient-based meta-learning that captures the imbalance of each user's rating distribution and accordingly computes adaptive loss for user-specific learning. It is the first work to tackle the impact of imbalanced ratings in cold-start sequential recommendation scenarios. We design adaptive weighted loss and improve the existing meta-learning algorithms for state-of-the-art sequential recommendation methods. Extensive experiments conducted on real-world datasets demonstrate the effectiveness of our framework.

Task-Agnostic Continual Reinforcement Learning: In Praise of a Simple Baseline

We study task-agnostic continual reinforcement learning (TACRL) in which standard RL challenges are compounded with partial observability stemming from task agnosticism, as well as additional difficulties of continual learning (CL), i.e., learning on a non-stationary sequence of tasks. Here we compare TACRL methods with their soft upper bounds prescribed by previous literature: multi-task learning (MTL) methods which do not have to deal with non-stationary data distributions, as well as task-aware methods, which are allowed to operate under full observability. We consider a previously unexplored and straightforward baseline for TACRL, replay-based recurrent RL (3RL), in which we augment an RL algorithm with recurrent mechanisms to address partial observability and experience replay mechanisms to address catastrophic forgetting in CL. Studying empirical performance in a sequence of RL tasks, we find surprising occurrences of 3RL matching and overcoming the MTL and task-aware soft upper bounds. We lay out hypotheses that could explain this inflection point of continual and task-agnostic learning research. Our hypotheses are empirically tested in continuous control tasks via a large-scale study of the popular multi-task and continual learning benchmark Meta-World. By analyzing different training statistics including gradient conflict, we find evidence that 3RL's outperformance stems from its ability to quickly infer how new tasks relate with the previous ones, enabling forward transfer.