FortisAVQA and MAVEN: a Benchmark Dataset and Debiasing Framework for Robust Multimodal Reasoning

Audio-Visual Question Answering (AVQA) is a challenging multimodal reasoning task requiring intelligent systems to answer natural language queries based on paired audio-video inputs accurately. However, existing AVQA approaches often suffer from overfitting to dataset biases, leading to poor robustness. Moreover, current datasets may not effectively diagnose these methods. To address these challenges, we first introduce a novel dataset, FortisAVQA, constructed in two stages: (1) rephrasing questions in the test split of the public MUSIC-AVQA dataset and (2) introducing distribution shifts across questions. The first stage expands the test space with greater diversity, while the second enables a refined robustness evaluation across rare, frequent, and overall question distributions. Second, we introduce a robust Multimodal Audio-Visual Epistemic Network (MAVEN) that leverages a multifaceted cycle collaborative debiasing strategy to mitigate bias learning. Experimental results demonstrate that our architecture achieves state-of-the-art performance on FortisAVQA, with a notable improvement of 7.81\%. Extensive ablation studies on both datasets validate the effectiveness of our debiasing components. Additionally, our evaluation reveals the limited robustness of existing multimodal QA methods. We also verify the plug-and-play capability of our strategy by integrating it with various baseline models across both datasets. Our dataset and code are available at https://github.com/reml-group/fortisavqa.

Multimodal Point Cloud Semantic Segmentation With Virtual Point Enhancement

LiDAR-based 3D point cloud recognition has been proven beneficial in various applications. However, the sparsity and varying density pose a significant challenge in capturing intricate details of objects, particularly for medium-range and small targets. Therefore, we propose a multi-modal point cloud semantic segmentation method based on Virtual Point Enhancement (VPE), which integrates virtual points generated from images to address these issues. These virtual points are dense but noisy, and directly incorporating them can increase computational burden and degrade performance. Therefore, we introduce a spatial difference-driven adaptive filtering module that selectively extracts valuable pseudo points from these virtual points based on density and distance, enhancing the density of medium-range targets. Subsequently, we propose a noise-robust sparse feature encoder that incorporates noise-robust feature extraction and fine-grained feature enhancement. Noise-robust feature extraction exploits the 2D image space to reduce the impact of noisy points, while fine-grained feature enhancement boosts sparse geometric features through inner-voxel neighborhood point aggregation and downsampled voxel aggregation. The results on the SemanticKITTI and nuScenes, two large-scale benchmark data sets, have validated effectiveness, significantly improving 2.89\% mIoU with the introduction of 7.7\% virtual points on nuScenes.

Single-Step Latent Consistency Model for Remote Sensing Image Super-Resolution

Recent advancements in diffusion models (DMs) have greatly advanced remote sensing image super-resolution (RSISR). However, their iterative sampling processes often result in slow inference speeds, limiting their application in real-time tasks. To address this challenge, we propose the latent consistency model for super-resolution (LCMSR), a novel single-step diffusion approach designed to enhance both efficiency and visual quality in RSISR tasks. Our proposal is structured into two distinct stages. In the first stage, we pretrain a residual autoencoder to encode the differential information between high-resolution (HR) and low-resolution (LR) images, transitioning the diffusion process into a latent space to reduce computational costs. The second stage focuses on consistency diffusion learning, which aims to learn the distribution of residual encodings in the latent space, conditioned on LR images. The consistency constraint enforces that predictions at any two timesteps along the reverse diffusion trajectory remain consistent, enabling direct mapping from noise to data. As a result, the proposed LCMSR reduces the iterative steps of traditional diffusion models from 50-1000 or more to just a single step, significantly improving efficiency. Experimental results demonstrate that LCMSR effectively balances efficiency and performance, achieving inference times comparable to non-diffusion models while maintaining high-quality output.

Dual-Domain Homogeneous Fusion with Cross-Modal Mamba and Progressive Decoder for 3D Object Detection

Fusing LiDAR point cloud features and image features in a homogeneous BEV space has been widely adopted for 3D object detection in autonomous driving. However, such methods are limited by the excessive compression of multi-modal features. While some works explore feature fusion in dense voxel spaces, they suffer from high computational costs and inefficiencies in query generation. To address these limitations, we propose a Dual-Domain Homogeneous Fusion network (DDHFusion), which leverages the complementary advantages of both BEV and voxel domains while mitigating their respective drawbacks. Specifically, we first transform image features into BEV and sparse voxel spaces using LSS and our proposed semantic-aware feature sampling module which can significantly reduces computational overhead by filtering unimportant voxels. For feature encoding, we design two networks for BEV and voxel feature fusion, incorporating novel cross-modal voxel and BEV Mamba blocks to resolve feature misalignment and enable efficient yet comprehensive scene perception. The output voxel features are injected into the BEV space to compensate for the loss of 3D details caused by height compression. For feature decoding, a progressive query generation module is implemented in the BEV domain to alleviate false negatives during query selection caused by feature compression and small object sizes. Finally, a progressive decoder can sequentially aggregate not only context-rich BEV features but also geometry-aware voxel features, ensuring more precise confidence prediction and bounding box regression. On the NuScenes dataset, DDHfusion achieves state-of-the-art performance, and further experiments demonstrate its superiority over other homogeneous fusion methods.

Positioning-Aided Channel Estimation for Multi-LEO Satellite Downlink Communications

We investigate a multi-low Earth orbit (LEO) satellite system that simultaneously provides positioning and communication services to terrestrial user terminals. To address the challenges of channel estimation in LEO satellite systems, we propose a novel two-timescale positioning-aided channel estimation framework, exploiting the distinct variation rates of position-related parameters and channel gains inherent in LEO satellite channels. Using the misspecified Cramer-Rao bound (MCRB) theory, we systematically analyze positioning performance under practical imperfections, such as inter-satellite clock bias and carrier frequency offset. Furthermore, we theoretically demonstrate how position information derived from downlink positioning can enhance uplink channel estimation accuracy, even in the presence of positioning errors, through an MCRB-based analysis. To overcome the constraints of limited link budgets and communication rates associated with single-satellite-based communication, we develop a distributed beamforming strategy for downlink communication. This strategy allows LEO satellites to independently optimize their beamformers using local channel state information, eliminating the need for centralized processing while preserving the advantages of multi-satellite cooperative communication. Theoretical analyses and numerical results confirm the effectiveness of the proposed framework in achieving high-precision downlink positioning under practical imperfections, facilitating uplink channel estimation, and enabling efficient downlink communication.

Integrated Positioning and Communication via LEO Satellites: Opportunities and Challenges

Low Earth orbit (LEO) satellites, as a prominent technology in the 6G non-terrestrial network, offer both positioning and communication capabilities. While these two applications have each been extensively studied and have achieved substantial progress in recent years, the potential synergistic benefits of integrating them remain an underexplored yet promising avenue. This article comprehensively analyzes the integrated positioning and communication (IPAC) systems on LEO satellites. By leveraging the distinct characteristics of LEO satellites, we examine how communication systems can enhance positioning accuracy and, conversely, how positioning information can be exploited to improve communication efficiency. In particular, we present two case studies to illustrate the potential of such integration. Finally, several key open research challenges in the LEO-based IPAC systems are discussed.

Latent Diffusion, Implicit Amplification: Efficient Continuous-Scale Super-Resolution for Remote Sensing Images

Recent advancements in diffusion models have significantly improved performance in super-resolution (SR) tasks. However, previous research often overlooks the fundamental differences between SR and general image generation. General image generation involves creating images from scratch, while SR focuses specifically on enhancing existing low-resolution (LR) images by adding typically missing high-frequency details. This oversight not only increases the training difficulty but also limits their inference efficiency. Furthermore, previous diffusion-based SR methods are typically trained and inferred at fixed integer scale factors, lacking flexibility to meet the needs of up-sampling with non-integer scale factors. To address these issues, this paper proposes an efficient and elastic diffusion-based SR model (E$^2$DiffSR), specially designed for continuous-scale SR in remote sensing imagery. E$^2$DiffSR employs a two-stage latent diffusion paradigm. During the first stage, an autoencoder is trained to capture the differential priors between high-resolution (HR) and LR images. The encoder intentionally ignores the existing LR content to alleviate the encoding burden, while the decoder introduces an SR branch equipped with a continuous scale upsampling module to accomplish the reconstruction under the guidance of the differential prior. In the second stage, a conditional diffusion model is learned within the latent space to predict the true differential prior encoding. Experimental results demonstrate that E$^2$DiffSR achieves superior objective metrics and visual quality compared to the state-of-the-art SR methods. Additionally, it reduces the inference time of diffusion-based SR methods to a level comparable to that of non-diffusion methods.

Attention-Guided Perturbation for Consistency Regularization in Semi-Supervised Medical Image Segmentation

Medical image segmentation is a pivotal step in diagnostic and therapeutic processes. However, the acquisition of high-quality annotated data is often constrained by scarcity and cost. Semi-supervised learning offers a promising approach to enhance model performance by using unlabeled data. While consistency regularization is a prevalent method in semi-supervised image segmentation, there is a dearth of research on perturbation strategies tailored for semi-supervised medical image segmentation tasks. This paper introduces an attention-guided perturbation strategy for semi-supervised consistency regularization in the context of medical image segmentation. We add the perturbation based on the attention from the model in the image and feature level to achieve consistency regularization. The method is adept at accommodating the intricate structures and high-dimensional semantics inherent in medical images, thereby enhancing the performance of semi-supervised segmentation tasks. Our method achieved state-of-the-art results on benchmark datasets, including a 90.4\% Dice score on the ACDC dataset in the 7-case scenario.

Unraveling and Mitigating Safety Alignment Degradation of Vision-Language Models

The safety alignment ability of Vision-Language Models (VLMs) is prone to be degraded by the integration of the vision module compared to its LLM backbone. We investigate this phenomenon, dubbed as ''safety alignment degradation'' in this paper, and show that the challenge arises from the representation gap that emerges when introducing vision modality to VLMs. In particular, we show that the representations of multi-modal inputs shift away from that of text-only inputs which represent the distribution that the LLM backbone is optimized for. At the same time, the safety alignment capabilities, initially developed within the textual embedding space, do not successfully transfer to this new multi-modal representation space. To reduce safety alignment degradation, we introduce Cross-Modality Representation Manipulation (CMRM), an inference time representation intervention method for recovering the safety alignment ability that is inherent in the LLM backbone of VLMs, while simultaneously preserving the functional capabilities of VLMs. The empirical results show that our framework significantly recovers the alignment ability that is inherited from the LLM backbone with minimal impact on the fluency and linguistic capabilities of pre-trained VLMs even without additional training. Specifically, the unsafe rate of LLaVA-7B on multi-modal input can be reduced from 61.53% to as low as 3.15% with only inference-time intervention. WARNING: This paper contains examples of toxic or harmful language.

Detecting Training Data of Large Language Models via Expectation Maximization

The widespread deployment of large language models (LLMs) has led to impressive advancements, yet information about their training data, a critical factor in their performance, remains undisclosed. Membership inference attacks (MIAs) aim to determine whether a specific instance was part of a target model's training data. MIAs can offer insights into LLM outputs and help detect and address concerns such as data contamination and compliance with privacy and copyright standards. However, applying MIAs to LLMs presents unique challenges due to the massive scale of pre-training data and the ambiguous nature of membership. Additionally, creating appropriate benchmarks to evaluate MIA methods is not straightforward, as training and test data distributions are often unknown. In this paper, we introduce EM-MIA, a novel MIA method for LLMs that iteratively refines membership scores and prefix scores via an expectation-maximization algorithm, leveraging the duality that the estimates of these scores can be improved by each other. Membership scores and prefix scores assess how each instance is likely to be a member and discriminative as a prefix, respectively. Our method achieves state-of-the-art results on the WikiMIA dataset. To further evaluate EM-MIA, we present OLMoMIA, a benchmark built from OLMo resources, which allows us to control the difficulty of MIA tasks with varying degrees of overlap between training and test data distributions. We believe that EM-MIA serves as a robust MIA method for LLMs and that OLMoMIA provides a valuable resource for comprehensively evaluating MIA approaches, thereby driving future research in this critical area.