arXiv subjects

Tianxiang Zhang

Publications and source records attributed to Tianxiang Zhang.

16 recordsLinked to original sources

arXiv2026

AC-LIO: Towards Asymptotic Compensation for Distortion in LiDAR-Inertial Odometry via Selective Intra-Frame Smoothing

Existing LiDAR-Inertial Odometry (LIO) methods typically utilize the prior trajectory derived from the IMU integration to compensate for the motion distortion within LiDAR frames. However, discrepancies between the prior and true trajectory can lead to residual motion distortions that compromise the consistency of LiDAR frame with its corresponding geometric environment. This imbalance may result in pointcloud registration becoming trapped in local optima, thereby exacerbating drift during long-term and large-scale localization. To this end, we propose a novel LIO framework with selective intra-frame smoothing dubbed AC-LIO. Our core idea is to asymptotically backpropagate current update term and compensate for residual motion distortion under the guidance of convergence criteria, aiming to improve the accuracy of discrete-state LIO system with minimal computational increase. Extensive experiments demonstrate that our AC-LIO framework further enhances odometry accuracy compared to prior arts, with about 30.4% reduction in average RMSE over the second best result, leading to marked improvements in the accuracy of long-term and large-scale localization and mapping.

cs.RO↗

arXiv2026

Reinforcing VLAs in Task-Agnostic World Models

Post-training Vision-Language-Action (VLA) models via reinforcement learning (RL) in learned world models has emerged as an effective strategy to adapt to new tasks without costly real-world interactions. However, while using imagined trajectories reduces the sample complexity of policy training, existing methods still heavily rely on task-specific data to fine-tune both the world and reward models, fundamentally limiting their scalability to unseen tasks. To overcome this, we argue that world and reward models should capture transferable physical priors that enable zero-shot inference. We propose RAW-Dream (Reinforcing VLAs in task-Agnostic World Dreams), a new paradigm that completely disentangles world model learning from downstream task dependencies. RAW-Dream utilizes a world model pre-trained on diverse task-free behaviors for predicting future rollouts, and an off-the-shelf Vision-Language Model (VLM) for reward generation. Because both components are task-agnostic, VLAs can be readily finetuned for any new task entirely within this zero-shot imagination. Furthermore, to mitigate world model hallucinations, we introduce a dual-noise verification mechanism to filter out unreliable rollouts. Extensive experiments across simulation and real-world settings demonstrate consistent performance gains, proving that generalized physical priors can effectively substitute for costly task-dependent data, offering a highly scalable roadmap for VLA adaptation.

cs.AI↗

arXiv2026

QA-ReID: Quality-Aware Query-Adaptive Convolution Leveraging Fused Global and Structural Cues for Clothes-Changing ReID

Unlike conventional person re-identification (ReID), clothes-changing ReID (CC-ReID) presents severe challenges due to substantial appearance variations introduced by clothing changes. In this work, we propose the Quality-Aware Dual-Branch Matching (QA-ReID), which jointly leverages RGB-based features and parsing-based representations to model both global appearance and clothing-invariant structural cues. These heterogeneous features are adaptively fused through a multi-modal attention module. At the matching stage, we further design the Quality-Aware Query Adaptive Convolution (QAConv-QA), which incorporates pixel-level importance weighting and bidirectional consistency constraints to enhance robustness against clothing variations. Extensive experiments demonstrate that QA-ReID achieves state-of-the-art performance on multiple benchmarks, including PRCC, LTCC, and VC-Clothes, and significantly outperforms existing approaches under cross-clothing scenarios.

cs.CV↗

arXiv2025

Discover, Learn, and Reinforce: Scaling Vision-Language-Action Pretraining with Diverse RL-Generated Trajectories

Scaling vision-language-action (VLA) model pre-training requires large volumes of diverse, high-quality manipulation trajectories. Most current data is obtained via human teleoperation, which is expensive and difficult to scale. Reinforcement learning (RL) methods learn useful skills through autonomous exploration, making them a viable approach for generating data. However, standard RL training collapses to a narrow execution pattern, limiting its utility for large-scale pre-training. We propose Discover, Lea rn and Reinforce (DLR), an information-theoretic pattern discovery framework that generates multiple distinct, high-success behavioral patterns for VLA pretraining. Empirically, DLR generates a markedly more diverse trajectory corpus on LIBERO. Specifically, it learns multiple distinct, high-success strategies for the same task where standard RL discovers only one, and hence it covers substantially broader regions of the state-action space. When adapted to unseen downstream task suites, VLA models pretrained on our diverse RL data surpass counterparts trained on equal-sized standard RL datasets. Moreover, DLR exhibits positive data-scaling behavior that single-pattern RL lacks. These results position multi-pattern RL as a practical, scalable data engine for embodied foundation models.

cs.RO↗

arXiv2025

Fine-Grained DINO Tuning with Dual Supervision for Face Forgery Detection

The proliferation of sophisticated deepfakes poses significant threats to information integrity. While DINOv2 shows promise for detection, existing fine-tuning approaches treat it as generic binary classification, overlooking distinct artifacts inherent to different deepfake methods. To address this, we propose a DeepFake Fine-Grained Adapter (DFF-Adapter) for DINOv2. Our method incorporates lightweight multi-head LoRA modules into every transformer block, enabling efficient backbone adaptation. DFF-Adapter simultaneously addresses authenticity detection and fine-grained manipulation type classification, where classifying forgery methods enhances artifact sensitivity. We introduce a shared branch propagating fine-grained manipulation cues to the authenticity head. This enables multi-task cooperative optimization, explicitly enhancing authenticity discrimination with manipulation-specific knowledge. Utilizing only 3.5M trainable parameters, our parameter-efficient approach achieves detection accuracy comparable to or even surpassing that of current complex state-of-the-art methods.

cs.CV↗

arXiv2025

Multi-Channel Amplitude-Phase Asymmetric-Encrypted Janus Acoustic Meta-Holograms

Encrypted optical and acoustic meta-holograms only focus on the encrypted hologram in a single channel, viz. modulating spatial amplitude to project a holographic image. In this research, the unique concept of multi-channel amplitude-phase asymmetric-encrypted Janus acoustic meta-holograms is proposed, demonstrating remarkable capabilities of generating, encrypting, and decrypting both amplitude and phase holographic images on both sides of a metascreen. The flexible and decoupled manipulation mechanism for the amplitude-phase of the bidirectional acoustic waves used in our concept offers multiple possibilities to apply various encryption methods. In this work, our system enables single-input, two-faced four-channel asymmetric encryption, which substantially increase the communication capacity of conventional acoustic holograms, and establish a security framework based on mathematical problem, proving its security. Our work can lead to concrete applications including, but not limited to, multi-channel acoustic field communications and acoustic illusion and cloaking in non-transparent media.

physics.app-ph↗

arXiv2025

Neural Brain: A Neuroscience-inspired Framework for Embodied Agents

The rapid evolution of artificial intelligence (AI) has shifted from static, data-driven models to dynamic systems capable of perceiving and interacting with real-world environments. Despite advancements in pattern recognition and symbolic reasoning, current AI systems, such as large language models, remain disembodied, unable to physically engage with the world. This limitation has driven the rise of embodied AI, where autonomous agents, such as humanoid robots, must navigate and manipulate unstructured environments with human-like adaptability. At the core of this challenge lies the concept of Neural Brain, a central intelligence system designed to drive embodied agents with human-like adaptability. A Neural Brain must seamlessly integrate multimodal sensing and perception with cognitive capabilities. Achieving this also requires an adaptive memory system and energy-efficient hardware-software co-design, enabling real-time action in dynamic environments. This paper introduces a unified framework for the Neural Brain of embodied agents, addressing two fundamental challenges: (1) defining the core components of Neural Brain and (2) bridging the gap between static AI models and the dynamic adaptability required for real-world deployment. To this end, we propose a biologically inspired architecture that integrates multimodal active sensing, perception-cognition-action function, neuroplasticity-based memory storage and updating, and neuromorphic hardware/software optimization. Furthermore, we also review the latest research on embodied agents across these four aspects and analyze the gap between current AI systems and human intelligence. By synthesizing insights from neuroscience, we outline a roadmap towards the development of generalizable, autonomous agents capable of human-level intelligence in real-world scenarios.

cs.RO↗

arXiv2025

Learning Like Humans: Advancing LLM Reasoning Capabilities via Adaptive Difficulty Curriculum Learning and Expert-Guided Self-Reformulation

Despite impressive progress in areas like mathematical reasoning, large language models still face significant challenges in consistently solving complex problems. Drawing inspiration from key human learning strategies, we propose two novel strategies to enhance the capability of large language models to solve these complex problems. First, Adaptive Difficulty Curriculum Learning (ADCL) is a novel curriculum learning strategy that tackles the Difficulty Shift phenomenon (i.e., a model's perception of problem difficulty dynamically changes during training) by periodically re-estimating difficulty within upcoming data batches to maintain alignment with the model's evolving capabilities. Second, Expert-Guided Self-Reformulation (EGSR) is a novel reinforcement learning strategy that bridges the gap between imitation learning and pure exploration by guiding models to reformulate expert solutions within their own conceptual framework, rather than relying on direct imitation, fostering deeper understanding and knowledge assimilation. Extensive experiments on challenging mathematical reasoning benchmarks, using Qwen2.5-7B as the base model, demonstrate that these human-inspired strategies synergistically and significantly enhance performance. Notably, their combined application improves performance over the standard Zero-RL baseline by 10% on the AIME24 benchmark and 16.6% on AIME25.

cs.AI↗

arXiv2025

MoL-RL: Distilling Multi-Step Environmental Feedback into LLMs for Feedback-Independent Reasoning

Large language models (LLMs) face significant challenges in effectively leveraging sequential environmental feedback (EF) signals, such as natural language evaluations, for feedback-independent chain-of-thought (CoT) reasoning. Existing approaches either convert EF into scalar rewards, losing rich contextual information, or employ refinement datasets, failing to exploit the multi-step and discrete nature of EF interactions. To address these limitations, we propose MoL-RL, a novel training paradigm that integrates multi-step EF signals into LLMs through a dual-objective optimization framework. Our method combines MoL (Mixture-of-Losses) continual training, which decouples domain-specific EF signals (optimized via cross-entropy loss) and general language capabilities (preserved via Kullback-Leibler divergence), with GRPO-based post-training to distill sequential EF interactions into single-step inferences. This synergy enables robust feedback-independent reasoning without relying on external feedback loops. Experimental results on mathematical reasoning (MATH-500, AIME24/AIME25) and code generation (CodeAgent-Test) benchmarks demonstrate that MoL-RL achieves state-of-the-art performance with the Qwen3-8B model, while maintaining strong generalization across model scales (Qwen3-4B). This work provides a promising approach for leveraging multi-step textual feedback to enhance LLMs' reasoning capabilities in diverse domains.

cs.CL↗

arXiv2025

Referring Remote Sensing Image Segmentation via Bidirectional Alignment Guided Joint Prediction

Referring Remote Sensing Image Segmentation (RRSIS) is critical for ecological monitoring, urban planning, and disaster management, requiring precise segmentation of objects in remote sensing imagery guided by textual descriptions. This task is uniquely challenging due to the considerable vision-language gap, the high spatial resolution and broad coverage of remote sensing imagery with diverse categories and small targets, and the presence of clustered, unclear targets with blurred edges. To tackle these issues, we propose \ours, a novel framework designed to bridge the vision-language gap, enhance multi-scale feature interaction, and improve fine-grained object differentiation. Specifically, \ours introduces: (1) the Bidirectional Spatial Correlation (BSC) for improved vision-language feature alignment, (2) the Target-Background TwinStream Decoder (T-BTD) for precise distinction between targets and non-targets, and (3) the Dual-Modal Object Learning Strategy (D-MOLS) for robust multimodal feature reconstruction. Extensive experiments on the benchmark datasets RefSegRS and RRSIS-D demonstrate that \ours achieves state-of-the-art performance. Specifically, \ours improves the overall IoU (oIoU) by 3.76 percentage points (80.57) and 1.44 percentage points (79.23) on the two datasets, respectively. Additionally, it outperforms previous methods in the mean IoU (mIoU) by 5.37 percentage points (67.95) and 1.84 percentage points (66.04), effectively addressing the core challenges of RRSIS with enhanced precision and robustness.

cs.CV↗

arXiv2024

A Real-time Degeneracy Sensing and Compensation Method for Enhanced LiDAR SLAM

LiDAR is widely used in Simultaneous Localization and Mapping (SLAM) and autonomous driving. The LiDAR odometry is of great importance in multi-sensor fusion. However, in some unstructured environments, the point cloud registration cannot constrain the poses of the LiDAR due to its sparse geometric features, which leads to the degeneracy of multi-sensor fusion accuracy. To address this problem, we propose a novel real-time approach to sense and compensate for the degeneracy of LiDAR. Firstly, this paper introduces the degeneracy factor with clear meaning, which can measure the degeneracy of LiDAR. Then, the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering method adaptively perceives the degeneracy with better environmental generalization. Finally, the degeneracy perception results are utilized to fuse LiDAR and IMU, thus effectively resisting degeneracy effects. Experiments on our dataset show the method's high accuracy and robustness and validate our algorithm's adaptability to different environments and LiDAR scanning modalities.

cs.RO↗

arXiv2024

AS-LIO: Spatial Overlap Guided Adaptive Sliding Window LiDAR-Inertial Odometry for Aggressive FOV Variation

LiDAR-Inertial Odometry (LIO) demonstrates outstanding accuracy and stability in general low-speed and smooth motion scenarios. However, in high-speed and intense motion scenarios, such as sharp turns, two primary challenges arise: firstly, due to the limitations of IMU frequency, the error in estimating significantly non-linear motion states escalates; secondly, drastic changes in the Field of View (FOV) may diminish the spatial overlap between LiDAR frame and pointcloud map (or between frames), leading to insufficient data association and constraint degradation. To address these issues, we propose a novel Adaptive Sliding window LIO framework (AS-LIO) guided by the Spatial Overlap Degree (SOD). Initially, we assess the SOD between the LiDAR frames and the registered map, directly evaluating the adverse impact of current FOV variation on pointcloud alignment. Subsequently, we design an adaptive sliding window to manage the continuous LiDAR stream and control state updates, dynamically adjusting the update step according to the SOD. This strategy enables our odometry to adaptively adopt higher update frequency to precisely characterize trajectory during aggressive FOV variation, thus effectively reducing the non-linear error in positioning. Meanwhile, the historical constraints within the sliding window reinforce the frame-to-map data association, ensuring the robustness of state estimation. Experiments show that our AS-LIO framework can quickly perceive and respond to challenging FOV change, outperforming other state-of-the-art LIO frameworks in terms of accuracy and robustness.

cs.RO↗

arXiv2024

Adaptive FSS: A Novel Few-Shot Segmentation Framework via Prototype Enhancement

The Few-Shot Segmentation (FSS) aims to accomplish the novel class segmentation task with a few annotated images. Current FSS research based on meta-learning focus on designing a complex interaction mechanism between the query and support feature. However, unlike humans who can rapidly learn new things from limited samples, the existing approach relies solely on fixed feature matching to tackle new tasks, lacking adaptability. In this paper, we propose a novel framework based on the adapter mechanism, namely Adaptive FSS, which can efficiently adapt the existing FSS model to the novel classes. In detail, we design the Prototype Adaptive Module (PAM), which utilizes accurate category information provided by the support set to derive class prototypes, enhancing class-specific information in the multi-stage representation. In addition, our approach is compatible with diverse FSS methods with different backbones by simply inserting PAM between the layers of the encoder. Experiments demonstrate that our method effectively improves the performance of the FSS models (e.g., MSANet, HDMNet, FPTrans, and DCAMA) and achieve new state-of-the-art (SOTA) results (i.e., 72.4\% and 79.1\% mIoU on PASCAL-5$^i$ 1-shot and 5-shot settings, 52.7\% and 60.0\% mIoU on COCO-20$^i$ 1-shot and 5-shot settings). Our code can be available at https://github.com/jingw193/AdaptiveFSS.

cs.CV↗

arXiv2023

Positive-Negative Equal Contrastive Loss for Semantic Segmentation

The contextual information is critical for various computer vision tasks, previous works commonly design plug-and-play modules and structural losses to effectively extract and aggregate the global context. These methods utilize fine-label to optimize the model but ignore that fine-trained features are also precious training resources, which can introduce preferable distribution to hard pixels (i.e., misclassified pixels). Inspired by contrastive learning in unsupervised paradigm, we apply the contrastive loss in a supervised manner and re-design the loss function to cast off the stereotype of unsupervised learning (e.g., imbalance of positives and negatives, confusion of anchors computing). To this end, we propose Positive-Negative Equal contrastive loss (PNE loss), which increases the latent impact of positive embedding on the anchor and treats the positive as well as negative sample pairs equally. The PNE loss can be directly plugged right into existing semantic segmentation frameworks and leads to excellent performance with neglectable extra computational costs. We utilize a number of classic segmentation methods (e.g., DeepLabV3, HRNetV2, OCRNet, UperNet) and backbone (e.g., ResNet, HRNet, Swin Transformer) to conduct comprehensive experiments and achieve state-of-the-art performance on three benchmark datasets (e.g., Cityscapes, COCO-Stuff and ADE20K). Our code will be publicly available soon.

cs.CV↗

arXiv2022

TransBTSV2: Towards Better and More Efficient Volumetric Segmentation of Medical Images

Transformer, benefiting from global (long-range) information modeling using self-attention mechanism, has been successful in natural language processing and computer vision recently. Convolutional Neural Networks, capable of capturing local features, are difficult to model explicit long-distance dependencies from global feature space. However, both local and global features are crucial for dense prediction tasks, especially for 3D medical image segmentation. In this paper, we present the further attempt to exploit Transformer in 3D CNN for 3D medical image volumetric segmentation and propose a novel network named TransBTSV2 based on the encoder-decoder structure. Different from TransBTS, the proposed TransBTSV2 is not limited to brain tumor segmentation (BTS) but focuses on general medical image segmentation, providing a stronger and more efficient 3D baseline for volumetric segmentation of medical images. As a hybrid CNN-Transformer architecture, TransBTSV2 can achieve accurate segmentation of medical images without any pre-training, possessing the strong inductive bias as CNNs and powerful global context modeling ability as Transformer. With the proposed insight to redesign the internal structure of Transformer block and the introduced Deformable Bottleneck Module to capture shape-aware local details, a highly efficient architecture is achieved with superior performance. Extensive experimental results on four medical image datasets (BraTS 2019, BraTS 2020, LiTS 2017 and KiTS 2019) demonstrate that TransBTSV2 achieves comparable or better results compared to the state-of-the-art methods for the segmentation of brain tumor, liver tumor as well as kidney tumor. Code will be publicly available at https://github.com/Wenxuan-1119/TransBTS.

eess.IV↗

arXiv2022

Attention guided global enhancement and local refinement network for semantic segmentation

The encoder-decoder architecture is widely used as a lightweight semantic segmentation network. However, it struggles with a limited performance compared to a well-designed Dilated-FCN model for two major problems. First, commonly used upsampling methods in the decoder such as interpolation and deconvolution suffer from a local receptive field, unable to encode global contexts. Second, low-level features may bring noises to the network decoder through skip connections for the inadequacy of semantic concepts in early encoder layers. To tackle these challenges, a Global Enhancement Method is proposed to aggregate global information from high-level feature maps and adaptively distribute them to different decoder layers, alleviating the shortage of global contexts in the upsampling process. Besides, a Local Refinement Module is developed by utilizing the decoder features as the semantic guidance to refine the noisy encoder features before the fusion of these two (the decoder features and the encoder features). Then, the two methods are integrated into a Context Fusion Block, and based on that, a novel Attention guided Global enhancement and Local refinement Network (AGLN) is elaborately designed. Extensive experiments on PASCAL Context, ADE20K, and PASCAL VOC 2012 datasets have demonstrated the effectiveness of the proposed approach. In particular, with a vanilla ResNet-101 backbone, AGLN achieves the state-of-the-art result (56.23% mean IoU) on the PASCAL Context dataset. The code is available at https://github.com/zhasen1996/AGLN.

cs.CV↗