39 accepted papers
Video generation has recently emerged as a central task in the field of generative AI. However, the substantial computational cost inherent in video synthesis makes model distillation a critical technique for efficient deployment. Despite its significance, there is a scarcity of methods specifically
Large Vision--Language Models (VLMs) unlearning tends to eliminate the influence of ``to-be-forgotten'' content in the training corpora, algorithmically by suppressing the likelihood of faithfully generating responses on forget-target inputs. The injection of adversarial inputs can manipulate the un
Large Language Models (LLMs) demonstrate significant advantages in leveraging structured world knowledge and multi-step reasoning capabilities. However, fundamental challenges arise when transforming LLMs into real-world recommendation systems due to semantic and behavioral misalignment. To bridge
Short-form videos have become a primary medium for digital advertising, requiring scalable and efficient content creation. However, current workflows and AI tools remain disjoint and modality-specific, leading to high production costs and low overall efficiency. To address this issue, we propose Aut
Experimental design is critical for evidence-based decision-making in healthcare, marketing, and public policy. However, designing efficient experiments across heterogeneous subgroups presents significant challenges. Existing methods often optimize for statistical power or overall sample efficiency,
Graph-based vertical federated learning (GVFL) enables multiple parties to collaboratively train and infer over aligned nodes, where each party contributes its own local embedding derived from different attributes and adjacency relations. Adversarial inputs injected by an attacker can skew the joint
Learning to rank can be cast as a probabilistic modeling problem over permutations, where the goal is to estimate the likelihood of an observed total ordering of items. This formulation naturally involves full-order probabilities of the form $\mathbb{P}(\mathrm{z}_1 < \cdots < \mathrm{z}_n)$, whose …
We present Narrative Weaver, a novel framework that addresses a fundamental challenge in generative AI: achieving controllable, long-range, and consistent visual content generation. While existing models excel at generating high-fidelity short-form visual content, they struggle to maintain narrative
Reinforcement learning (RL) has equipped LLM agents with a strong ability to solve complex tasks. However, existing RL methods normally use a single policy network, causing simplicity bias where simple tasks occupy most parameters and dominate gradient updates, leaving insufficient capacity for comp…
We present Stylos, a single-forward 3D Gaussian framework for 3D style transfer that operates on unposed content, from a single image to a multi-view collection, conditioned on a separate reference style image. Stylos synthesizes a stylized 3D Gaussian scene without per-scene optimization or precomp…
Recent progress in text-to-image (T2I) diffusion models (DMs) has enabled high-quality visual synthesis from diverse textual prompts. Yet, most existing T2I DMs, even those equipped with large language model (LLM)-based text encoders, remain text-pixel mappers -- they employ LLMs merely as text enco…
Videos showcasing specific products are increasingly important for E-commerce. Key moments naturally exist as the first appearance of a specific product, presentation of its distinctive features, the presence of a buying link, etc. Adding proper sound effects (SFX) to such moments, or video decorati…
Vertical federated learning (VFL) enables feature-level collaboration by incorporating scattered attributes from aligned samples, and allows each party to contribute its personalized input to joint training and inference. The injection of adversarial inputs can mislead the joint inference towards th
Large language models (LLMs) develop the in-context learning capability through pretraining and instruction tuning, enabling task adaptation without parameter updates. Self-refinement is a manifestation of this capability, which allows LLMs to iteratively refine the output using self-generated feedb
Contemporary recommendation systems predominantly rely on ID embedding to capture latent associations among users and items. However, this approach overlooks the wealth of semantic information embedded within textual descriptions of items, leading to suboptimal performance and poor generalizations.…
User simulators can rapidly generate a large volume of timely user behavior data, providing a testing platform for reinforcement learning-based recommender systems, thus accelerating their iteration and optimization. However, prevalent user simulators generally suffer from significant limitations, i…
Cascade Ranking is a prevalent architecture in large-scale top-k selection systems like recommendation and advertising platforms. Traditional training methods focus on single-stage optimization, neglecting interactions between stages. Recent advances have introduced interaction-aware training paradi…
In many machine learning tasks, it is often necessary for the relationship between input and output variables to be monotonic, including both strictly monotonic and implicitly monotonic relationships. Traditional methods for maintaining monotonicity mainly rely on construction or regularization tech…
Recommender systems are increasingly prevalent to provide personalized suggestions and enhance user satisfaction. Typical recommendation models encode users and items as embeddings, and generate recommendations by assessing the similarity between these embeddings. Despite their effectiveness, these…
Adding proper background music helps complete a short video to be shared. Previous work tackles the task by video-to-music retrieval (V2MR), aiming to find the most suitable music track from a collection to match the content of a given query video. In practice, however, music tracks are typically mu…
We introduce Orthus, a unified multimodal model that excels in generating interleaved images and text from mixed-modality inputs by simultaneously handling discrete text tokens and continuous image features under the \textbf{AR} modeling principle. The continuous treatment of visual signals minimize…
Road space and lane line segmentation are vital tasks in intelligent traffic surveillance, yet they receive less attention compared to similar tasks in autonomous driving. In highway surveillance, segmenting occluded road areas and lane lines is crucial for comprehensive perception, though it adds c…
Multimodal recommendation aims to integrate collaborative signals with heterogeneous content such as visual and textual information, but remains challenged by modality-specific noise, semantic inconsistency, and unstable propagation over user–item graphs. These issues are often exacerbated by naive…
This paper presents the Semantic-aWarE spatial-tEmporal Tokenizer (SweetTok), a novel video tokenizer to overcome the limitations in current video tokenization methods for compacted yet effective discretization. Unlike previous approaches that process flattened local visual patches via direct discre…
Cervical cytologic whole slide image (WSI) multiple classificaton (grading) is a challenging task. Current studies typically ignore the unbalanced data distribution and require multi-class annotations to learn cell features for WSI grading, which largely suffers from label noise. In this paper, we d…
Centimeter-scale underwater robots have important applications in underwater resource exploration, environmental monitoring, equipment fault diagnosis, and military applications. However, developing small underwater robots remains a challenge because of the limitation of miniaturized structures. In
In order to address the issue that previous deep learning methods for classifying pathological images cannot adaptively learn features, we propose an end-to-end attention pooling method based on a multi-instance learning patch scoring model. Our method integrates feature extraction and classificatio…
Convolutional Neural Networks (CNNs) have become the de facto approach for medical image classification in recent years. However, the deficiency of convolutional operations in extracting global features has limited the further improvement of this task. Vision Transformers (ViTs) can model long-range…
Breast cancer histopathological image classification has made great progress with the use of Convolutional Neural Networks (CNNs). However, due to the limited receptive field, CNNs have difficulty in learning the global information of breast cancer histopathological images, hindering the further imp…
Long-term engagement is preferred over immediate engagement in sequential recommendation as it directly affects product operational metrics such as daily active users (DAUs) and dwell time. Meanwhile, reinforcement learning (RL) is widely regarded as a promising framework for optimizing long-term en…
In many real-world scenarios, Reinforcement Learning (RL) algorithms are trained on data with dynamics shift, i.e., with different underlying environment dynamics. A majority of current methods address such issue by training context encoders to identify environment parameters. Data with dynamics shi…
Sparse training is a popular technique to reduce the overhead of training large models. Although previous work has shown promising results for nonstructured sparse models, it is still unclear whether a sparse model with structural constraints can be trained from scratch to high accuracy. In this wor…
Pre-trained models have proved to be powerful in enhancing task-oriented dialog systems. However, current pre-training methods mainly focus on enhancing dialog understanding and generation tasks while neglecting the exploitation of dialog policy. In this paper, we propose GALAXY, a novel pre-trained…
We present a boundary-aware domain adaptation model for LiDAR scan full-scene semantic segmentation (LiDARNet). Our model can extract both the domain private features and the domain shared features with a two branch structure. We embedded Gated-SCNN into the segmentor component of LiDARNet to learn…
Semantic scene understanding is crucial for robust and safe autonomous navigation, particularly so in off-road environments. Recent deep learning advances for 3D semantic segmentation rely heavily on large sets of training data, however existing autonomy datasets either represent urban environments…
Scribble-supervised semantic segmentation has gained much attention recently for its promising performance without high-quality annotations. Due to the lack of supervision, confident and consistent predictions are usually hard to obtain. Typically, people handle these problems by either adopting an…
The large communication overhead has imposed a bottleneck on the performance of distributed Stochastic Gradient Descent (SGD) for training deep neural networks. Previous works have demonstrated the potential of using gradient sparsification and quantization to reduce the communication cost. Howeve…
Deep Neural Networks (DNNs) have recently shown state of the art performance on semantic segmentation tasks, however, they still suffer from problems of poor boundary localization and spatial fragmented predictions. The difficulties lie in the requirement of making dense predictions from a long path…
In this work, we propose a generic scheme to promote any diffusion-based salient object detection algorithm by original ways to re-synthesize the diffusion matrix and construct the seed vector. We first make a novel analysis of the working mechanism of the diffusion matrix, which reveals the close r…