Xinyu Xiao

Sen Lei, Xinyu Xiao, Tianlin Zhang et al.

Given a language expression, referring remote sensing image segmentation (RRSIS) aims to identify ground objects and assign pixel-wise labels within the imagery. The one of key challenges for this task is to capture discriminative multi-modal features via text-image alignment. However, the existing RRSIS methods use one vanilla and coarse alignment, where the language expression is directly extracted to be fused with the visual features. In this paper, we argue that a ``fine-grained image-text alignment'' can improve the extraction of multi-modal information. To this point, we propose a new referring remote sensing image segmentation method to fully exploit the visual and linguistic representations. Specifically, the original referring expression is regarded as context text, which is further decoupled into the ground object and spatial position texts. The proposed fine-grained image-text alignment module (FIAM) would simultaneously leverage the features of the input image and the corresponding texts, obtaining better discriminative multi-modal representation. Meanwhile, to handle the various scales of ground objects in remote sensing, we introduce a Text-aware Multi-scale Enhancement Module (TMEM) to adaptively perform cross-scale fusion and intersections. We evaluate the effectiveness of the proposed method on two public referring remote sensing datasets including RefSegRS and RRSIS-D, and our method obtains superior performance over several state-of-the-art methods. The code will be publicly available at https://github.com/Shaosifan/FIANet.

Shaohong Wang, Lu Bin, Xinyu Xiao et al.

Multi-agent collaborative perception has emerged as a widely recognized technology in the field of autonomous driving in recent years. However, current collaborative perception predominantly relies on LiDAR point clouds, with significantly less attention given to methods using camera images. This severely impedes the development of budget-constrained collaborative systems and the exploitation of the advantages offered by the camera modality. This work proposes an instance-level fusion transformer for visual collaborative perception (IFTR), which enhances the detection performance of camera-only collaborative perception systems through the communication and sharing of visual features. To capture the visual information from multiple agents, we design an instance feature aggregation that interacts with the visual features of individual agents using predefined grid-shaped bird eye view (BEV) queries, generating more comprehensive and accurate BEV features. Additionally, we devise a cross-domain query adaptation as a heuristic to fuse 2D priors, implicitly encoding the candidate positions of targets. Furthermore, IFTR optimizes communication efficiency by sending instance-level features, achieving an optimal performance-bandwidth trade-off. We evaluate the proposed IFTR on a real dataset, DAIR-V2X, and two simulated datasets, OPV2V and V2XSet, achieving performance improvements of 57.96%, 9.23% and 12.99% in AP@70 metrics compared to the previous SOTAs, respectively. Extensive experiments demonstrate the superiority of IFTR and the effectiveness of its key components. The code is available at https://github.com/wangsh0111/IFTR.

Dingkun Zhang, Shuhan Qi, Yulin Wu et al.

Multimodal Large Language Models (MLLMs) suffer from severe training inefficiency issue, which is associated with their massive model sizes and visual token numbers. Existing efforts in efficient training focus on reducing model sizes or trainable parameters. Inspired by the success of Visual Token Pruning (VTP) in improving inference efficiency, we are exploring another substantial research direction for efficient training by reducing visual tokens. However, applying VTP at the training stage results in a training-inference mismatch: pruning-trained models perform poorly when inferring on non-pruned full visual token sequences. To close this gap, we propose DualSpeed, a fast-slow framework for efficient training of MLLMs. The fast-mode is the primary mode, which incorporates existing VTP methods as plugins to reduce visual tokens, along with a mode isolator to isolate the model's behaviors. The slow-mode is the auxiliary mode, where the model is trained on full visual sequences to retain training-inference consistency. To boost its training, it further leverages self-distillation to learn from the sufficiently trained fast-mode. Together, DualSpeed can achieve both training efficiency and non-degraded performance. Experiments show DualSpeed accelerates the training of LLaVA-1.5 by 2.1$\times$ and LLaVA-NeXT by 4.0$\times$, retaining over 99% performance. Code: https://github.com/dingkun-zhang/DualSpeed

Wenwen Zhang, Xinyu Xiao, Hangguan Shan et al.

One-shot object detection (OSOD) aims to detect all object instances towards the given category specified by a query image. Most existing studies in OSOD endeavor to explore effective cross-image correlation and alleviate the semantic feature misalignment, however, ignoring the phenomenon of the model bias towards the base classes and the generalization degradation on the novel classes. Observing this, we propose a novel framework, namely Base-class Suppression and Prior Guidance (BSPG) network to overcome the problem. Specifically, the objects of base categories can be explicitly detected by a base-class predictor and adaptively eliminated by our base-class suppression module. Moreover, a prior guidance module is designed to calculate the correlation of high-level features in a non-parametric manner, producing a class-agnostic prior map to provide the target features with rich semantic cues and guide the subsequent detection process. Equipped with the proposed two modules, we endow the model with a strong discriminative ability to distinguish the target objects from distractors belonging to the base classes. Extensive experiments show that our method outperforms the previous techniques by a large margin and achieves new state-of-the-art performance under various evaluation settings.

Inclusion AI, Biao Gong, Cheng Zou et al.

We propose Ming-Omni, a unified multimodal model capable of processing images, text, audio, and video, while demonstrating strong proficiency in both speech and image generation. Ming-Omni employs dedicated encoders to extract tokens from different modalities, which are then processed by Ling, an MoE architecture equipped with newly proposed modality-specific routers. This design enables a single model to efficiently process and fuse multimodal inputs within a unified framework, thereby facilitating diverse tasks without requiring separate models, task-specific fine-tuning, or structural redesign. Importantly, Ming-Omni extends beyond conventional multimodal models by supporting audio and image generation. This is achieved through the integration of an advanced audio decoder for natural-sounding speech and Ming-Lite-Uni for high-quality image generation, which also allow the model to engage in context-aware chatting, perform text-to-speech conversion, and conduct versatile image editing. Our experimental results showcase Ming-Omni offers a powerful solution for unified perception and generation across all modalities. Notably, our proposed Ming-Omni is the first open-source model we are aware of to match GPT-4o in modality support, and we release all code and model weights to encourage further research and development in the community.

Inclusion AI, Biao Gong, Cheng Zou et al.

We introduce Ming-Lite-Uni, an open-source multimodal framework featuring a newly designed unified visual generator and a native multimodal autoregressive model tailored for unifying vision and language. Specifically, this project provides an open-source implementation of the integrated MetaQueries and M2-omni framework, while introducing the novel multi-scale learnable tokens and multi-scale representation alignment strategy. By leveraging a fixed MLLM and a learnable diffusion model, Ming-Lite-Uni enables native multimodal AR models to perform both text-to-image generation and instruction based image editing tasks, expanding their capabilities beyond pure visual understanding. Our experimental results demonstrate the strong performance of Ming-Lite-Uni and illustrate the impressive fluid nature of its interactive process. All code and model weights are open-sourced to foster further exploration within the community. Notably, this work aligns with concurrent multimodal AI milestones - such as ChatGPT-4o with native image generation updated in March 25, 2025 - underscoring the broader significance of unified models like Ming-Lite-Uni on the path toward AGI. Ming-Lite-Uni is in alpha stage and will soon be further refined.

Xinyu Xiao, Sen Lei, Eryun Liu et al.

Precipitation nowcasting is vital for flood warning, agricultural management, and emergency response, yet two bottlenecks persist: the prohibitive cost of modeling million-scale spatiotemporal tokens from multi-variate atmospheric fields, and the extreme long-tailed rainfall distribution where heavy-to-torrential events -- those of greatest societal impact -- constitute fewer than 0.1% of all samples. We propose the Precipitation-Adaptive Network (PA-Net), a Transformer framework whose computational budget is explicitly governed by rainfall intensity. Its core component, Precipitation-Adaptive MoE (PA-MoE), dynamically scales the number of activated experts per token according to local precipitation magnitude, channeling richer representational capacity toward the rare yet critical heavy-rainfall tail. A Dual-Axis Compressed Latent Attention mechanism factorizes spatiotemporal attention with convolutional reduction to manage massive context lengths, while an intensity-aware training protocol progressively amplifies learning signals from extreme-rainfall samples. Experiment on ERA5 demonstrate consistent improvements over state-of-the-art baselines, with particularly significant gains in heavy-rain and rainstorm regimes.

Xinyu Xiao, Binbin Yang, Tingtian Li et al.

Diffusion-based text-to-video generation (T2V) or image-to-video (I2V) generation have emerged as a prominent research focus. However, there exists a challenge in integrating the two generative paradigms into a unified model. In this paper, we present a unified video generation model (UniVid) with hybrid conditions of the text prompt and reference image. Given these two available controls, our model can extract objects' appearance and their motion descriptions from textual prompts, while obtaining texture details and structural information from image clues to guide the video generation process. Specifically, we scale up the pre-trained text-to-image diffusion model for generating temporally coherent frames via introducing our temporal-pyramid cross-frame spatial-temporal attention modules and convolutions. To support bimodal control, we introduce a dual-stream cross-attention mechanism, whose attention scores can be freely re-weighted for interpolation of between single and two modalities controls during inference. Extensive experiments showcase that our UniVid achieves superior temporal coherence on T2V, I2V and (T+I)2V tasks.

Qizhao Jin, Xianhuang Xu, Yong Cao et al.

Recently, Transformer-based architectures have advanced meteorological prediction. However, this position-centric tokenizer conflicts with the core principle of meteorological systems, where the weather phenomena undoubtedly involve synergistic interactions among multiple elements while positional information constitutes merely a component of the boundary conditions. This paper focuses primarily on the task of precipitation nowcasting and develops an efficient distribution-centric Meteorological Tokenization (MeTok) scheme, which spatially sequences to group similar meteorological features. Based on the rearrangement, realigned group learning enhances robustness across precipitation patterns, especially extreme ones. Specifically, we introduce the Hyper-Aligned Grouping Transformer (HyAGTransformer) with two key improvements: 1) The Grouping Attention (GA) mechanism uses MeTok to enable self-aligned learning of features from different precipitation patterns; 2) The Neighborhood Feed-Forward Network (N-FFN) integrates adjacent group features, aggregating contextual information to boost patch embedding discriminability. Experiments on the ERA5 dataset for 6-hour forecasts show our method improves the IoU metric by at least 8.2% in extreme precipitation prediction compared to other methods. Additionally, it gains performance with more training data and increased parameters, demonstrating scalability, stability, and superiority over traditional methods.

Shaohong Wang, Bin Lu, Xinyu Xiao et al.

Collaborative visual perception methods have gained widespread attention in the autonomous driving community in recent years due to their ability to address sensor limitation problems. However, the absence of explicit depth information often makes it difficult for camera-based perception systems, e.g., 3D object detection, to generate accurate predictions. To alleviate the ambiguity in depth estimation, we propose RayFusion, a ray-based fusion method for collaborative visual perception. Using ray occupancy information from collaborators, RayFusion reduces redundancy and false positive predictions along camera rays, enhancing the detection performance of purely camera-based collaborative perception systems. Comprehensive experiments show that our method consistently outperforms existing state-of-the-art models, substantially advancing the performance of collaborative visual perception. The code is available at https://github.com/wangsh0111/RayFusion.

Inclusion AI, Bowen Ma, Cheng Zou et al.

We propose Ming-Flash-Omni, an upgraded version of Ming-Omni, built upon a sparser Mixture-of-Experts (MoE) variant of Ling-Flash-2.0 with 100 billion total parameters, of which only 6.1 billion are active per token. This architecture enables highly efficient scaling (dramatically improving computational efficiency while significantly expanding model capacity) and empowers stronger unified multimodal intelligence across vision, speech, and language, representing a key step toward Artificial General Intelligence (AGI). Compared to its predecessor, the upgraded version exhibits substantial improvements across multimodal understanding and generation. We significantly advance speech recognition capabilities, achieving state-of-the-art performance in contextual ASR and highly competitive results in dialect-aware ASR. In image generation, Ming-Flash-Omni introduces high-fidelity text rendering and demonstrates marked gains in scene consistency and identity preservation during image editing. Furthermore, Ming-Flash-Omni introduces generative segmentation, a capability that not only achieves strong standalone segmentation performance but also enhances spatial control in image generation and improves editing consistency. Notably, Ming-Flash-Omni achieves state-of-the-art results in text-to-image generation and generative segmentation, and sets new records on all 12 contextual ASR benchmarks, all within a single unified architecture.

Dingkun Zhang, Shuhan Qi, Xinyu Xiao et al.

Recent advances in Multimodal Large Language Models (MLLMs) have enhanced their versatility as they integrate a growing number of modalities. Considering the heavy cost of training MLLMs, it is efficient to reuse the existing ones and extend them to more modalities through Modality-incremental Continual Learning (MCL). The exploration of MCL is in its early stages. In this work, we dive into the causes of performance degradation in MCL. We uncover that it suffers not only from forgetting as in traditional continual learning, but also from misalignment between the modality-agnostic and modality-specific components. To this end, we propose an elegantly simple MCL paradigm called "MErge then ReAlign" (MERA) to address both forgetting and misalignment. MERA avoids introducing heavy model budgets or modifying model architectures, hence is easy to deploy and highly reusable in the MLLM community. Extensive experiments demonstrate the impressive performance of MERA, holding an average of 99.84\% Backward Relative Gain when extending to four modalities, achieving nearly lossless MCL performance. Our findings underscore the misalignment issue in MCL. More broadly, our work showcases how to adjust different components of MLLMs during continual learning.

Tingtian Li, Zixun Sun, Xinyu Xiao

Identifying highlight moments of raw video materials is crucial for improving the efficiency of editing videos that are pervasive on internet platforms. However, the extensive work of manually labeling footage has created obstacles to applying supervised methods to videos of unseen categories. The absence of an audio modality that contains valuable cues for highlight detection in many videos also makes it difficult to use multimodal strategies. In this paper, we propose a novel model with cross-modal perception for unsupervised highlight detection. The proposed model learns representations with visual-audio level semantics from image-audio pair data via a self-reconstruction task. To achieve unsupervised highlight detection, we investigate the latent representations of the network and propose the representation activation sequence learning (RASL) module with k-point contrastive learning to learn significant representation activations. To connect the visual modality with the audio modality, we use the symmetric contrastive learning (SCL) module to learn the paired visual and audio representations. Furthermore, an auxiliary task of masked feature vector sequence (FVS) reconstruction is simultaneously conducted during pretraining for representation enhancement. During inference, the cross-modal pretrained model can generate representations with paired visual-audio semantics given only the visual modality. The RASL module is used to output the highlight scores. The experimental results show that the proposed framework achieves superior performance compared to other state-of-the-art approaches.

Yukang Chen, Tong Yang, Xiangyu Zhang et al.

Object detectors are usually equipped with backbone networks designed for image classification. It might be sub-optimal because of the gap between the tasks of image classification and object detection. In this work, we present DetNAS to use Neural Architecture Search (NAS) for the design of better backbones for object detection. It is non-trivial because detection training typically needs ImageNet pre-training while NAS systems require accuracies on the target detection task as supervisory signals. Based on the technique of one-shot supernet, which contains all possible networks in the search space, we propose a framework for backbone search on object detection. We train the supernet under the typical detector training schedule: ImageNet pre-training and detection fine-tuning. Then, the architecture search is performed on the trained supernet, using the detection task as the guidance. This framework makes NAS on backbones very efficient. In experiments, we show the effectiveness of DetNAS on various detectors, for instance, one-stage RetinaNet and the two-stage FPN. We empirically find that networks searched on object detection shows consistent superiority compared to those searched on ImageNet classification. The resulting architecture achieves superior performance than hand-crafted networks on COCO with much less FLOPs complexity.