AAAI 2025 论文列表

Cross-domain recommendations in healthcare services differ from traditional ones in electronic commerce due to the need for heightened medical privacy protection for a small group of users, while ensuring the majority, who may lack sufficient medical knowledge, can understand the recommendations. To recommend doctors who provide online consultations to health video viewers and enable multimodal cross-domain recommendations from short video platforms (source domain) to online healthcare communities (target domain), this paper introduces a framework based on the User-Centric Synthetic Data Architect (UCSDA) and Pre-trained Large Language Model (PtLLM). UCSDA employs a user-centric, advanced selection-synthesis mechanism to filter users' cold interaction items and synthesize noise items, reducing privacy leakage risk. PtLLM focuses on necessary patient and doctor IDs during the recommendation decision process to generate explanations. The model's effectiveness and scalability were validated using three public datasets and a healthcare cross-domain recommendation dataset. In addition to traditional evaluation metrics, strong privacy metrics and the unique sentence ratio were used to assess privacy protection and interpretability. We also compared the characteristics of privacy protection and interpretability between e-commerce and healthcare recommendation scenarios.

Generalized few-shot semantic segmentation (GFSS) aims to segment objects of both base and novel classes, using sufficient samples of base classes and few samples of novel classes. Representative GFSS approaches typically employ a two-phase training scheme, involving base class pre-training followed by novel class fine-tuning, to learn the classifiers for base and novel classes respectively. Nevertheless, distribution gap exists between base and novel classes in this process. To narrow this gap, we exploit effective knowledge transfer from base to novel classes. First, a novel prototype modulation module is designed to modulate novel class prototypes by exploiting the correlations between base and novel classes. Second, a novel classifier calibration module is proposed to calibrate the weight distribution of the novel classifier according to that of the base classifier. Furthermore, existing GFSS approaches suffer from a lack of contextual information for novel classes due to their limited samples, we thereby introduce a context consistency learning scheme to transfer the contextual knowledge from base to novel classes. Extensive experiments on PASCAL-5i and COCO-20i demonstrate that our approach significantly enhances the state of the art in the GFSS setting.

3D occupancy perception accurately estimates the volumetric status and semantic labels of a scene, attracting significant attention in the field of autonomous driving. However, enhancing the model's ability to generalize across different driving scenarios or sensing systems, often requires redesigning the model or extra-expensive annotations. To this end, following a comprehensive analysis of the occupancy model architecture, we proposed the UGOCC method that utilizes domain adaptation to efficiently harness unlabeled autonomous driving data, thereby enhancing the model's generalizability. Specifically, we design the depth fusion module by employing self-supervised depth estimation, and propose a strategy based on semantic attention and domain adversarial learning to improve the generalizability of the learnable fusion module. Additionally, we propose an OCC-specific pseudo-label selection tailored for semi-supervised learning, which optimizes the overall network's generalizability. Our experiment results on two challenging datasets nuScenes and Waymo, demonstrate that our method not only achieves state-of-the-art generalizability but also enhances the model's perceptual capabilities within the source domain by utilizing unlabeled data.

Diffusion-based generative models have recently excelled in generating molecular conformations but struggled with the generalization issue -- models trained on one dataset may produce meaningless conformations on out-of-distribution molecules. On the other hand, distance geometry serves as a generalizable tool for the traditional computational chemistry methods of molecular conformation, which is predicated on the assumption that it is possible to adequately define the set of all potential conformations of any non-rigid molecular system using purely geometric constraints. In this work, we for the first time explicitly incorporate distance geometry constraints into pretraining phase of diffusion-based molecular generation models to improve the generalizability. Inspired by the classical distance geometry solution designed for solving the molecular distance geometry problem, we propose MiGDiff, a Metrization-Informed Geometric Diffusion framework. MiGDiff injects distance geometry constraints by pretraining the deep geometric diffusion backbone within the Metrization sampling approach, yielding a "Metrization-driven pretraining + Data-driven finetuning" paradigm. Experimental results demonstrate that MiGDiff outperforms state-of-the-art methods and possesses strong generalization capabilities, particularly on generating previously unseen molecules, revealing the vast untapped potential of combining traditional computational methods with deep generative models for 3D molecular generation.

Existing Vision-Language Pretraining (VLP) methods have achieved remarkable improvements across a variety of vision-language tasks, confirming their effectiveness in capturing coarse-grained semantic correlations. However, their capability for fine-grained understanding, which is critical for many nuanced vision-language applications, remains limited. Prevailing VLP models often overlook the intricate distinctions in expressing different modal features and typically depend on the similarity of holistic features for cross-modal interactions. Moreover, these models directly align and integrate features from different modalities, focusing more on coarse-grained general representations, thus failing to capture the nuanced differences necessary for tasks demanding a more detailed perception. In response to these limitations, we introduce Negative Augmented Samples(NAS), a refined vision-language pretraining model that innovatively incorporates NAS to specifically address the challenge of fine-grained understanding. NAS utilizes a Visual Dictionary(VD) as a semantic bridge between visual and linguistic domains. Additionally, it employs a Negative Visual Augmentation(NVA) method based on the VD to generate challenging negative image samples. These samples deviate from positive samples exclusively at the token level, thereby necessitating that the model discerns the subtle disparities between positive and negative samples with greater precision. Comprehensive experiments validate the efficacy of NAS components and underscore its potential to enhance fine-grained vision-language comprehension.

We address the challenging task of neural machine translation (NMT) in the entertainment domain, where the objective is to automatically translate a given dialogue from a source language content to a target language. This task has various applications, particularly in automatic dubbing, subtitling, and other content localization tasks, enabling source content to reach a wider audience. Traditional NMT systems typically translate individual sentences in isolation, without facilitating knowledge transfer of crucial elements such as the context and style from previously encountered sentences. In this work, we emphasize the significance of these fundamental aspects in producing pertinent and captivating translations. We demonstrate their significance through several examples and propose a novel framework for entertainment translation, which, to our knowledge, is the first of its kind. Furthermore, we introduce an algorithm to estimate the context and style of the current session and use these estimations to generate a prompt that guides a Large Language Model (LLM) to generate high-quality translations. Our method is both language and LLM-agnostic, making it a general-purpose tool. We demonstrate the effectiveness of our algorithm through various numerical studies and observe significant improvement in the COMET scores over various state-of-the-art LLMs. Moreover, our proposed method consistently outperforms baseline LLMs in terms of win-ratio.

Causal language models acquire vast amount of knowledge from general text corpus during pretraining, but the efficiency of knowledge learning is known to be unsatisfactory, especially when learning from knowledge-dense and small-sized corpora. The deficiency can come from long-distance dependencies which are hard to capture by language models, and overfitting to co-occurrence patterns and distracting clues in the training text. To address these issues, the paper proposes a method to enhance knowledge learning during language model pretraining, by enhancing elusive but important clues in text discovered by the language model themselves. We found that larger language models pay more attention to non-obvious but important clues, which are often overlooked by smaller language models. Therefore, we can identify these clues by contrasting the attention weights of large and small language models. We use the identified clues as a guide to perform token-dropout data augmentation on the training text, and observed a significant boost in both small and large models' performance in fact memorization. This shows that the behavior contrast between more and less-performant language models contains important clues for knowledge learning, and it can be "amplified" for a straight-forward improvement in knowledge learning efficiency.

Sequential recommendation aims to capture the temporal dependencies of items in a user's historical interactions and make recommendations based on this. Previous generative methods addressed the issue of data not directly reflecting user preference uncertainty by modeling the distribution of latent item representations. Diffusion model (DM)-based methods have achieved significant success due to their high-quality generation and stable training. However, they lack satisfactory user sequence representations to guide the generation process, impacting recommendation performance. Moreover, these methods overlook the drawback of slow inference speed, severely limiting their practical value. To obtain effective generative guidance signals and accelerate the recommendation process, we propose DAE4Rec. In this approach, a Graph Auto-Encoder (GAE) is used to obtain interpretable item node representations, revealing global transitions of items that previous methods struggled to uncover. Then, we use it to construct a generative guidance signal with lower coupling and variance for the diffusion model. Additionally, by employing a non-Markov chain derived from the forward diffusion process, it is the first to implement a 'skip-step' reverse process in diffusion model-based methods. And a creatively designed compensator is used to bridge the performance gap caused by 'skip-step'. Extensive experiments on three real-world datasets demonstrate that DAE4Rec outperforms other state-of-the-art generative sequential recommenders.

Intimate Partner Violence is a global, life-threatening public health issue that can be prevented by recognizing emotionally aggressive behaviors that signal the potential for future relationship abuse. To help identify these precursory unhealthy behaviors, this study proposes a Multi-task Learning framework for training robust models capable of detecting not only physically abusive behaviors but also emotionally abusive behaviors, such as belittling or manipulation, which historically precede physical abuse. Preliminary results indicate that Multi-task Learning can improve detection of emotional abuse and help tune detection models to particular kinds of relationship abuse.

Agents significantly enhance the capabilities of standalone Large Language Models (LLMs) by perceiving environments, making decisions, and executing actions. However, LLM agents still face challenges in tasks that require multiple decision-making steps. Estimating the value of actions in specific tasks is difficult when intermediate actions are neither appropriately rewarded nor penalized. In this paper, we propose leveraging a task-relevant Q-value model to guide action selection. Specifically, we first collect decision-making trajectories annotated with step-level Q values via Monte Carlo Tree Search (MCTS) and construct preference data. We then use another LLM to fit these preferences through step-level Direct Policy Optimization (DPO), which serves as the Q-value model. During inference, at each decision-making step, LLM agents select the action with the highest Q value before interacting with the environment. We apply our method to various open-source and API-based LLM agents, demonstrating that Q-value models significantly improve their performance. Notably, the performance of the agent built with Phi-3-mini-4k-instruct improved by 103% on WebShop and 75% on HotPotQA when enhanced with Q-value models, even surpassing GPT-4o-mini. Additionally, Q-value models offer several advantages, such as generalization to different LLM agents and seamless integration with existing prompting strategies.

Many critical business and societal decisions in areas such as supply chain and healthcare involve numerous potential actions, complex constraints, and goals that can be modeled as objective functions. Mathematical optimization, a core area in Operations Research (OR), provides robust, mathematically grounded methodologies to address such decisions and has shown tremendous benefits in many applications. However, its application requires the creation of accurate and efficient optimization models, necessitating rare expertise and considerable time, creating a barrier to widespread adoption in decision-making. Thus, it is a long-standing goal to make these capabilities widely accessible. The advent of Large Language Models (LLMs) has made advanced Artificial Intelligence (AI) capabilities widely accessible through natural language. LLMs can accelerate expert work in creating formal models like computer programs, and emerging research indicates they can also speed up the development of optimization models by OR experts. We, therefore, propose integrating and advancing LLM and optimization modeling to empower organizational decision-makers to model and solve such complex problems without requiring deep expertise in optimization. In this work, we present our vision for democratizing optimization modeling for organizational decision-making by such a combination of LLMs and optimization modeling. We identify a set of fundamental requirements for the vision's implementation and describe the state of the art through a literature survey and some experimentation. We show that a) LLMs already provide substantial novel capabilities relevant to realizing this vision, but that b) major research challenges remain to be addressed. We also propose possible research directions to overcome these gaps. We would like this work to serve as a call to action to bring together the LLM and OR optimization modeling communities to pursue this vision, thereby enabling much more widespread improved decision-making and increasing by orders of magnitude the benefits AI and OR can bring to enterprises and society.

Contrastive learning is a prevalent technique in self-supervised vision representation learning, typically generating positive pairs by applying two data augmentations to the same image. Designing effective data augmentation strategies is crucial for the success of contrastive learning. Inspired by the story of the blind men and the elephant, we introduce JointCrop and JointBlur. These methods generate more challenging positive pairs by leveraging the joint distribution of the two augmentation parameters, thereby enabling contrastive learning to acquire more effective feature representations. To the best of our knowledge, this is the first effort to explicitly incorporate the joint distribution of two data augmentation parameters into contrastive learning. As a plug-and-play framework without additional computational overhead, JointCrop and JointBlur enhance the performance of SimCLR, BYOL, MoCo v1, MoCo v2, MoCo v3, SimSiam, and Dino baselines with notable improvements.

Recent methods, such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS), have demonstrated remarkable capabilities in novel view synthesis. However, despite their success in producing high-quality images for viewpoints similar to those seen during training, they struggle when generating detailed images from viewpoints that significantly deviate from the training set, particularly in close-up views. The primary challenge stems from the lack of specific training data for close-up views, leading to the inability of current methods to render these views accurately. To address this issue, we introduce a novel pseudo-label-based learning strategy. This approach leverages pseudo-labels derived from existing training data to provide targeted supervision across a wide range of close-up viewpoints. Recognizing the absence of benchmarks for this specific challenge, we also present a new dataset designed to assess the effectiveness of both current and future methods in this area. Our extensive experiments demonstrate the efficacy of our approach.

Chain of Thought (CoT) prompting can encourage language models to engage in multi-step logical reasoning. The quality of the provided demonstrations significantly influences the success of downstream inference tasks. Current unsupervised CoT methods primarily select examples based on the semantics of the questions, which can introduce noise and lack interpretability. In this paper, we propose leveraging reasoning patterns to enhance CoT prompting effectiveness. Reasoning patterns represent the process by which language models arrive at their final results. By utilizing prior knowledge and prompt-based methods from large models, we first construct task-specific pattern sets. We then select diverse demonstrations based on different reasoning patterns. This approach not only mitigates the impact of noise but also provides explicit interpretability to help us understand the mechanisms of CoT. Extensive experiments demonstrate that our method is more robust and consistently leads to improvements across various reasoning tasks.

Audiovisual Automatic Speech Recognition (AV-ASR) aims to improve speech recognition accuracy by leveraging visual signals. It is particularly challenging in unconstrained real-world scenarios across various domains due to noisy acoustic environments, spontaneous speech, and the uncertain use of visual information. Most previous works fine-tune audio-only ASR models on audiovisual datasets, optimizing them for conventional ASR objectives. However, they often neglect visual features and common errors in unconstrained video scenarios. In this paper, we propose using a preference optimization strategy to improve speech recognition accuracy for real-world videos. First, we create preference data via simulating common errors that occurred in AV-ASR from two focals: manipulating the audio or vision input and rewriting the output transcript. Second, we propose BPO-AVASR, a Bifocal Preference Optimization method to improve AV-ASR models by leveraging both input-side and output-side preference. Extensive experiments demonstrate that our approach significantly improves speech recognition accuracy across various domains, outperforming previous state-of-the-art models on real-world video speech recognition.

Aging biomarkers play a crucial role in uncovering the biological mechanisms behind aging and in developing strategies to support healthy aging. However, the search for reliable aging biomarkers is particularly challenging due to the intricate and multifactorial nature of the aging process. Furthermore, biomarker names and categories are not well-standardized in the current literature. While, a formal definition of a biomarker is nonexistent in the current literature, formally defining biomarkers and standardizing the vocabulary for biomarkers can help accelerate AI research around this concept which can lead to better, faster and more accurate analyses of the existing data and literature. Thus, in this work, we generated Knowledge Graphs that can help us define and standardize biomarkers. We present our Knowledge Graphs (KGs) generated using both an LLM and expert-curated datasets. We compare both KGs to understand why systematic integration between these two models is needed. The integration of Knowledge Graphs (KGs) and Large Language Models (LLMs) presents a promising approach to advancing aging biomarker research through the inherent structured and standardized nature of ontology schemas in knowledge graphs. We showcase that the accuracy of LLM-generated KGs remains questionable but systematic methods such as KNARM can help us with the accuracy of these efforts. In future work, we will propose a synergistic framework where KGs and LLMs interact iteratively to improve both the comprehensiveness and accuracy of aging biomarker information.

Deep neural networks (DNNs) are vulnerable to adversarial examples (AEs) that mislead the model while appearing benign to human observers. A critical concern is the transferability of AEs, which enables black-box attacks without direct access to the target model. However, many previous attacks have failed to explain the intrinsic mechanism of adversarial transferability, lacking a unified and representative metric for transferability as well. In this paper, we rethink the property of transferable AEs and develop a novel metric to measure transferability from the perspective of generalization. Building on insights from this metric, we analyze the generalization of AEs across models with different architectures and prove that we can find a local perturbation to mitigate the gap between surrogate and target models. We further establish the inner connections between model smoothness and flat local maxima, both of which contribute to the transferability of AEs. Further, we propose a new adversarial attack algorithm, Adversarial Weight Tuning (AWT), which adaptively adjusts the parameters of the surrogate model using generated AEs to optimize the flat local maxima and model smoothness simultaneously, without the need for extra data. AWT is a data-free tuning method that combines gradient-based and model-related attack methods to enhance the transferability of AEs. Extensive experiments on a variety of models with different architectures on ImageNet demonstrate that AWT yields superior performance over other attacks, with an average increase of nearly 5% and 10% attack success rates on CNN-based and Transformer-based models, respectively, compared to state-of-the-art attacks.

We propose a novel sample selection method for image classification in the presence of noisy labels. Existing methods typically consider small-loss samples as correctly labeled. However, some correctly labeled samples are inherently difficult for the model to learn and can exhibit high loss similar to mislabeled samples in the early stages of training. Consequently, setting a threshold on per-sample loss to select correct labels results in a trade-off between precision and recall in sample selection: a lower threshold may miss many correctly labeled hard-to-learn samples (low recall), while a higher threshold may include many mislabeled samples (low precision). To address this issue, our goal is to accurately distinguish correctly labeled yet hard-to-learn samples from mislabeled ones, thus alleviating the trade-off dilemma. We achieve this by considering the trends in model prediction confidence rather than relying solely on loss values. Empirical observations show that only for correctly labeled samples, the model's prediction confidence for the annotated labels typically increases faster than for any other classes. Based on this insight, we propose tracking the confidence gaps between the annotated labels and other classes during training and evaluating their trends using the Mann-Kendall Test. A sample is considered potentially correctly labeled if all its confidence gaps tend to increase. Our method functions as a plug-and-play component that can be seamlessly integrated into existing sample selection techniques. Experiments on several standard benchmarks and real-world datasets demonstrate that our method enhances the performance of existing methods for learning with noisy labels.

Importance sampling is a rare event simulation technique used in Monte Carlo simulations to bias the sampling distribution towards the rare event of interest. By assigning appropriate weights to sampled points, importance sampling allows for more efficient estimation of rare events or tails of distributions. However, importance sampling can fail when the proposal distribution does not effectively cover the target distribution. In this work, we propose a method for more efficient sampling by updating the proposal distribution in the latent space of a normalizing flow. Normalizing flows learn an invertible mapping from a target distribution to a simpler latent distribution. The latent space can be more easily explored during the search for a proposal distribution, and samples from the proposal distribution are recovered in the space of the target distribution via the invertible mapping. We empirically validate our methodology on simulated robotics applications such as autonomous racing and aircraft ground collision avoidance.