AAAI 2025 论文列表

Image super-resolution (SR) is essential for bridging the gap between modern hardware and real-time computer graphics (CG) applications. It reduces CG workload by allowing low-resolution rendering, with original quality restored later via mathematical operations or machine learning. However, recent learning-based SR methods often rely on complex models, demanding high computational resources and undermining the benefits of reduced rendering workload. Our qualitative and quantitative analysis of the SR process and rendering reveals that readily accessible rendering information can significantly enhance neural network design by serving as additional features. To capitalize on this, we propose CGSR, an optimization framework designed for lightweight real-time super-resolution. CGSR utilizes rendering information to boost both network extensibility and efficiency. It utilizes progressively available rendering information from the pipeline, which arrives earlier than the rendered frame, enabling pre-processing and masking of latency. These features are then integrated into a selected SR network backbone to form a CG-enhanced network. This network is further optimized and refined into a CG-optimized version using neural architecture search (NAS). To improve runtime performance, CGSR also employs rendering-aware hybrid pruning, which dynamically prunes the network based on temporal rendering data. Evaluation results show that CGSR significantly reduces parameter size, multi-add operations, and inference time while maintaining high SR quality across various backbone SR networks.

Model ensembling is a widely used technique that enhances performance in image-text matching tasks by combining multiple models, each trained with different initializations. However, the inefficiencies associated with training several models and generating outputs from them constrain their practical applicability. In this paper, we argue that while the parameters of two randomly initialized models can differ significantly, their feature distributions can be similar at certain stages. By employing a proposed technique called cross-modal realignment, we demonstrate that features derived from differently initialized models maintain similarity at the feature extraction stage and can be effectively transformed by fine-tuning a small number of parameters. These findings provide an efficient way to achieve ensemble-like performance within a single model. Specifically, we propose a Feature Diversification Framework (FDF) that emulates the outputs of multiple model initializations to generate diverse features from a common shared feature. Firstly, we introduce feature conversion methods to transform shared features into a set of distinct features. Next, a realignment training strategy is presented to optimize negative pairs for realigning these transformed features, thereby enhancing their diversification to resemble the outputs of different models. Additionally, we propose a reweighting module that assigns weights to these features, enabling a weighted fusion approach for robust feature representation. Extensive experiments on the Flickr30K and MS-COCO datasets demonstrate the effectiveness and generalizability of our framework.

Student placements under diversity constraints are a common practice globally. This paper addresses the selection of students by a single school under a one-to-one convention, where students can belong to multiple types but are counted only once based on one type. While existing algorithms in economics and computer science aim to help schools meet diversity goals and priorities, we demonstrate that these methods can result in significant imbalances among students with different type combinations. To address this issue, we introduce a new property called balanced representation, which ensures fair representation across all types and type combinations. We propose a straightforward choice function that uniquely satisfies four fundamental properties: maximal diversity, non-wastefulness, justified envy-freeness, and balanced representation. While previous research has primarily focused on algorithms based on bipartite graphs, we take a different approach by utilizing flow networks. This method provides a more compact formalization of the problem and significantly improves computational efficiency. Additionally, we present efficient algorithms for implementing our choice function within both the bipartite graph and flow network frameworks.

Question answering on free-form tables (a.k.a. TableQA) is a challenging task because of the flexible structure and complex schema of tables. Recent studies use Large Language Models (LLMs) for this task, exploiting their capability in understanding the questions and tabular data, which are typically given in natural language and contain many textual fields, respectively. While this approach has shown promising results, it overlooks the challenges brought by numerical values which are common in tabular data, and LLMs are known to struggle with such values. We aim to address this issue, and we propose a model named TabLaP that uses LLMs as a planner rather than an answer generator. This approach exploits LLMs' capability in multi-step reasoning while leaving the actual numerical calculations to a Python interpreter for accurate calculation. Recognizing the inaccurate nature of LLMs, we further make a first attempt to quantify the trustworthiness of the answers produced by TabLaP, such that users can use TabLaP in a regret-aware manner. Experimental results on two benchmark datasets show that TabLaP is substantially more accurate than the state-of-the-art models, improving the answer accuracy by 5.7% and 5.8% on the two datasets, respectively.

Protein-nucleic acid interactions play a fundamental and critical role in a wide range of life activities. Accurate identification of nucleic acid-binding residues helps to understand the intrinsic mechanisms of the interactions. However, the accuracy and interpretability of existing computational methods for recognizing nucleic acid-binding residues need to be further improved. Here, we propose a novel method called GeSite based the domain-adaptive protein language model and E(3)-equivariant graph neural network. Prediction results across multiple benchmark test sets demonstrate that GeSite is superior or comparable to state-of-the-art prediction methods. The MCC values of GeSite are 0.522 and 0.326 for the one DNA-binding residue test set and one RNA-binding resi-due test set, which are 0.57 and 38.14% higher than that of the second-best method, respectively. Detailed experi-mental results suggest that the advanced performance of GeSite lies in the well-designed nucleic acid-binding pro-tein adaptive language model. Additionally, interpretabil-ity analysis exposes the perception of the prediction mod-el on various remote and close functional domains, which is the source of its discernment ability.

Given an edge-incomplete graph, how can we accurately find its missing links? The problem aims to discover the missing relations between entities when their relationships are represented as a graph. Edge-incomplete graphs are prevalent in real-world due to practical limitations, such as not checking all users when adding friends in a social network. Addressing the problem is crucial for various tasks, including recommending friends in social networks and finding references in citation networks. However, previous approaches rely heavily on the given edge-incomplete (observed) graph, making it challenging to consider the missing (unobserved) links. In this paper, we propose PULL, an accurate link prediction method based on the positive-unlabeled (PU) learning. PULL treats the observed edges in the training graph as positive examples, and the unconnected node pairs as unlabeled ones. PULL effectively prevents the link predictor from blindly trusting the observed graph by proposing latent variables for every edge, and leveraging the expected graph structure with respect to these variables. Extensive experiments on real- world datasets show that PULL consistently outperforms the baselines for predicting links in edge-incomplete graphs.

In this paper, we consider a perturbation-based metric of predictive faithfulness of feature rankings (or attributions) that we call PGI squared When applied to decision tree-based regression models, the metric can be computed exactly and efficiently for arbitrary independent feature perturbation distributions. In particular, the computation does not involve Monte Carlo sampling that has been typically used for computing similar metrics and which is inherently prone to inaccuracies. As a second contribution, we proposed a procedure for constructing feature ranking based on PGI squared. Our results indicate the proposed ranking method is comparable to the widely recognized SHAP explainer, offering a viable alternative for assessing feature importance in tree-based models.

Masked autoencoders employ random masking to effectively reconstruct input images using self-supervised techniques, which allows for efficient training on large datasets. However, the random masking strategy does not adequately tap into information encapsulated within high-dimensional hyperspectral satellite imagery that is used in several domains. We propose a novel masking strategy, HOGMAE, based on the Histogram of Oriented Gradients that incorporates rich information inherent within satellite images during the mask creation step. Our experiments, over a hyperspectral satellite dataset, demonstrate the effectiveness of our methodology.

Limited accessibility to neurological care leads to under-diagnosed Parkinson's Disease (PD), preventing early intervention. Existing AI-based PD detection methods primarily focus on unimodal analysis of motor or speech tasks, overlooking the multifaceted nature of the disease. To address this, we introduce a large-scale, multi-task video dataset of 1102 sessions (each containing videos of finger tapping, facial expression, and speech tasks captured via webcam) from 845 participants (272 with PD). We propose a novel Uncertainty-calibrated Fusion Network (UFNet) that leverages this multimodal data to enhance diagnostic accuracy. UFNet employs independent task-specific networks, trained with Monte Carlo Dropout for uncertainty quantification, followed by self-attended fusion of features, with attention weights dynamically adjusted based on task-specific uncertainties. We randomly split the participants into training (60%), validation (20%), and test (20%) sets to ensure patient-centered evaluation. UFNet significantly outperformed single-task models in terms of accuracy, area under the ROC curve (AUROC), and sensitivity while maintaining non-inferior specificity. Withholding uncertain predictions further boosted the performance, achieving 88.0 +- 0.3% accuracy, 93.0 +- 0.2% AUROC, 79.3 +- 0.9% sensitivity, and 92.6 +- 0.3% specificity, at the expense of not being able to predict for 2.3 +- 0.3% data (+- denotes 95% confidence interval). Further analysis suggests that the trained model does not exhibit any detectable bias across sex and ethnic subgroups and is most effective for individuals aged between 50 and 80. By merely requiring a webcam and microphone, our approach facilitates accessible home-based PD screening, especially in regions with limited healthcare resources.

We propose a 2D simulation system for multi-agent collective construction (MACC) based on simple line-following intelligent machines (SLIM) - small differential drive mobile robots. Our MACC-SLIM system alleviates the high upfront cost of implementing MACC on real hardware. Our system builds upon widely available resources, namely a standard LCD screen and commodity mobile robots, allowing researchers and schools easier access to MACC hardware implementation. We test the system on plans generated by an optimal state-of-the-art MACC algorithm, demonstrating there are still non-insignificant synchronization delays. The MACC-SLIM system allows us to observe bottlenecks, parallelism, and possible execution failures of plans generated by the MACC algorithms.

In recent years, as data and problem sizes have increased, distributed learning has become an essential tool for training high-performance models. However, the communication bottleneck, especially for high-dimensional data, is a challenge. Several techniques have been developed to overcome this problem. These include communication compression and implementation of local steps, which work particularly well when there is similarity of local data samples. In this paper, we study the synergy of these approaches for efficient distributed optimization. We propose the first theoretically grounded accelerated algorithms utilizing unbiased and biased compression under data similarity, leveraging variance reduction and error feedback frameworks. In terms of communication time our theory gives ?(1+[M^(-¼)+?^(-½)]√(? /?)) complexity for unbiased compressors and ?(1+β^(¼)√(? /?)) for biased ones, where M is the number of computational nodes, β is the compression power, ? is the similarity measure and ? is the parameter of strong convexity of the objective. Our theoretical results are of record and confirmed by experiments on different average losses and datasets.

Pan-sharpening aims to preserve the spectral information of the multi-spectral (MS) image while leveraging the high-frequency details from the guided high-resolution panchromatic (PAN) image to enhance its spatial resolution. The key challenge is how to preserve the spectral information from the MS image and the spatial details from the PAN image as much as possible. Diffusion models have achieved favorable results in image restoration and synthesis tasks but suffer from excessive computational resource and time consumption. In this paper, we design a novel and computationally efficient diffusion-based pan-sharpening network that achieves accelerated diffusion while reducing task complexity by decoupling the high and low-frequency components of the fused image. Specifically, leveraging the information-preserving characteristic of the wavelet transformation, we introduce a Wavelet-based Low-frequency Diffusion Model (WLDM). WLDM generates the low-frequency coefficient of high-resolution MS (HRMS) image from the low-resolution MS (LRMS) image. This approach significantly reduces computational resources and complexity compared to the direct restoration of the HRMS image. Furthermore, we have devised a High-frequency Information Restoration Module (HIRM) to restore the high-frequency information in the HRMS image through the interaction of high-frequency coefficients from the PAN image in three directions. Extensive experiments on three different datasets demonstrate that our method outperforms existing approaches in both quantitative metrics, qualitative metrics, and inference efficiency.

All-Weather Image Restoration (AWIR) under adverse weather conditions is a challenging task due to the presence of different types of degradations. Prior research in this domain relies on extensive training data but lacks the utilization of additional contextual information for restoration guidance. Consequently, the performance of existing methods is limited by the degradation cues that are learnt from individual training samples. Recent advancements in visual in-context learning have introduced generalist models that are capable of addressing multiple computer vision tasks simultaneously by using the information present in the provided context as a prior. In this paper, we propose All-Weather Image Restoration using Visual In-Context Learning (AWRaCLe), a novel approach for AWIR that innovatively utilizes degradation-specific visual context information to steer the image restoration process. To achieve this, AWRaCLe incorporates Degradation Context Extraction (DCE) and Context Fusion (CF) to seamlessly integrate degradation-specific features from the context into an image restoration network. The proposed DCE and CF blocks leverage CLIP features and incorporate attention mechanisms to adeptly learn and fuse contextual information. These blocks are specifically designed for visual in-context learning under all-weather conditions and are crucial for effective context utilization. Through extensive experiments, we demonstrate the effectiveness of AWRaCLe for all-weather restoration and show that our method advances the state-of-the-art in AWIR.

Adversarial attacks poses a significant threat to the security of AI-based systems. To counteract these attacks, adversarial training (AT) and ensemble learning (EL) have emerged as widely adopted methods for enhancing model robustness. However, a counter-intuitive phenomenon arises where the simple combination of these approaches may potentially compromising adversarial robustness of ensemble models. In this paper, we propose a novel method called Alignment and Unlearning for Training Ensembles (AUTE), aiming to effectively integrate AT and EL to maximize their benefits. Specifically, AUTE incorporates two key components. Firstly, AUTE divides the ensemble into a big peer model and a single member in a loop manner, aligning their outputs for boosting robustness of each member. Secondly, AUTE introduces the concept of unlearning, actively forgetting specific data with over-confident properties to preserve model capacity to learn more robust features. Extensive experiments across various datasets and networks illustrate that AUTE achieves superior performance compared to baselines. For instance, a 5-member AUTE with ResNet-20 networks outperforms state-of-the-art method by 2.1% and 3.2% in classifying clean and adversarial data. Additionally, AUTE can easily extend to non-adversarial training paradigm, surpassing current standard ensemble learning methods by a large margin.

Reinforcement learning is a widely used approach for training an agent to maximize rewards in a given environment. Action policies learned with this technique see a broad range of applications in practical areas like games, healthcare, robotics, or autonomous driving. However, enforcing ethical behavior or norms based on deontic constraints that the agent should adhere to during policy execution remains a complex challenge. Especially constraints that emerge after the training can necessitate to redo policy learning, which can be costly and, more critically, time-intense. In order to mitigate this problem, we present a framework for policy fixing in case of a norm violation, which allows the agent to stay operational. Based on answer set programming (ASP), emergency plans are generated that exclude or minimize cost of norm violations by future actions in a horizon of interest. By combining and developing optimization techniques, efficient policy fixing under real-time constraints can be achieved.

Quantization stands as a pivotal technique for large language model (LLM) serving, yet it poses significant challenges particularly in achieving effective low-bit quantization. The limited numerical mapping makes the quantized model produce a non-trivial error, bringing out intolerable performance degration. This paper is anchored in the basic idea of model compression objectives, and delves into the layer-wise error distribution of LLMs during post-training quantization. Subsequently, we introduce ASER, an algorithm consisting of (1) Error Reconstruction: low-rank compensation for quantization error with LoRA-style matrices constructed by whitening SVD; (2) Activation Smoothing: outlier extraction to gain smooth activation and better error compensation. ASER is capable of quantizing typical LLMs to low-bit ones, particularly preserving accuracy even in W4A8 per-channel setup. Experimental results show that ASER is competitive among the state-of-the-art quantization algorithms, showing potential to activation quantization, with minor overhead.

3D object detection in point clouds is critical in 3D computer vision, autonomous driving, and robotics. Existing point-based detectors, tailored to handle unstructured raw point clouds, often rely on simplistic sampling strategies to select a subset of points for local representation learning and detection. However, the diverse patterns exhibited by multiple types of point cloud data present a significant challenge to the universality of current detectors, particularly those captured by varied sensors (e.g., LiDAR and 4D Imaging Radar). In response to this challenge, we introduce an adaptable point-based single-stage 3D detector, AS-Det, engineered to excel on both LiDAR and 4D Radar point clouds. Specifically, we propose a novel active sampling strategy that actively mines object-related information to achieve efficient sampling and representation across different types of point clouds through end-to-end training. Additionally, we introduce a lightweight multi-scale center feature aggregation module to exploit multi-scale object context for precise and low-cost detection. By integrating the abovementioned modules, AS-Det achieves highly adaptive detection on various point clouds, encompassing different sensors and scales. Experimental results demonstrate the superior performance and adaptability of AS-Det on both LiDAR and 4D Radar point clouds.

Ensuring annotator quality in training and evaluation data is a key piece of machine learning in NLP. Tasks such as sentiment analysis and offensive speech detection are intrinsically subjective, creating a challenging scenario for traditional quality assessment approaches because it is hard to distinguish disagreement due to poor work from that due to differences of opinions between sincere annotators. With the goal of increasing diverse perspectives in annotation while ensuring consistency, we propose ARTICLE, an in-context learning (ICL) framework to estimate annotation quality through self-consistency. We evaluate this framework on two offensive speech datasets using multiple LLMs and compare its performance with traditional methods. Our findings indicate that ARTICLE can be used as a robust method for identifying reliable annotators, hence improving data quality.

Ensuring annotator quality in training and evaluation data is a key piece of machine learning in NLP. Tasks such as sentiment analysis and offensive speech detection are intrinsically subjective, creating a challenging scenario for traditional quality assessment approaches because it is hard to distinguish disagreement due to poor work from that due to differences of opinions between sincere annotators. With the goal of increasing diverse perspectives in annotation while ensuring consistency, we propose ARTICLE, an in-context learning (ICL) framework to estimate annotation quality through self-consistency. We evaluate this framework on two offensive speech datasets using multiple LLMs and compare its performance with traditional methods. Our findings indicate that ARTICLE can be used as a robust method for identifying reliable annotators, hence improving data quality.