Long-VLA: Unleashing Long-Horizon Capability of Vision Language Action Model for Robot Manipulation

Vision-Language-Action (VLA) models have become a cornerstone in robotic policy learning, leveraging large-scale multimodal data for robust and scalable control. However, existing VLA frameworks primarily address short-horizon tasks, and their effectiveness on long-horizon, multi-step robotic manipulation remains limited due to challenges in skill chaining and subtask dependencies. In this work, we introduce Long-VLA, the first end-to-end VLA model specifically designed for long-horizon robotic tasks. Our approach features a novel phase-aware input masking strategy that adaptively segments each subtask into moving and interaction phases, enabling the model to focus on phase-relevant sensory cues and enhancing subtask compatibility. This unified strategy preserves the scalability and data efficiency of VLA training, and our architecture-agnostic module can be seamlessly integrated into existing VLA models. We further propose the L-CALVIN benchmark to systematically evaluate long-horizon manipulation. Extensive experiments on both simulated and real-world tasks demonstrate that Long-VLA significantly outperforms prior state-of-the-art methods, establishing a new baseline for long-horizon robotic control.

Gradient Rectification for Robust Calibration under Distribution Shift

Deep neural networks often produce overconfident predictions, undermining their reliability in safety-critical applications. This miscalibration is further exacerbated under distribution shift, where test data deviates from the training distribution due to environmental or acquisition changes. While existing approaches improve calibration through training-time regularization or post-hoc adjustment, their reliance on access to or simulation of target domains limits their practicality in real-world scenarios. In this paper, we propose a novel calibration framework that operates without access to target domain information. From a frequency-domain perspective, we identify that distribution shifts often distort high-frequency visual cues exploited by deep models, and introduce a low-frequency filtering strategy to encourage reliance on domain-invariant features. However, such information loss may degrade In-Distribution (ID) calibration performance. Therefore, we further propose a gradient-based rectification mechanism that enforces ID calibration as a hard constraint during optimization. Experiments on synthetic and real-world shifted datasets, including CIFAR-10/100-C and WILDS, demonstrate that our method significantly improves calibration under distribution shift while maintaining strong in-distribution performance.

Enhancing Homophily-Heterophily Separation: Relation-Aware Learning in Heterogeneous Graphs

Real-world networks usually have a property of node heterophily, that is, the connected nodes usually have different features or different labels. This heterophily issue has been extensively studied in homogeneous graphs but remains under-explored in heterogeneous graphs, where there are multiple types of nodes and edges. Capturing node heterophily in heterogeneous graphs is very challenging since both node/edge heterogeneity and node heterophily should be carefully taken into consideration. Existing methods typically convert heterogeneous graphs into homogeneous ones to learn node heterophily, which will inevitably lose the potential heterophily conveyed by heterogeneous relations. To bridge this gap, we propose Relation-Aware Separation of Homophily and Heterophily (RASH), a novel contrastive learning framework that explicitly models high-order semantics of heterogeneous interactions and adaptively separates homophilic and heterophilic patterns. Particularly, RASH introduces dual heterogeneous hypergraphs to encode multi-relational bipartite subgraphs and dynamically constructs homophilic graphs and heterophilic graphs based on relation importance. A multi-relation contrastive loss is designed to align heterogeneous and homophilic/heterophilic views by maximizing mutual information. In this way, RASH simultaneously resolves the challenges of heterogeneity and heterophily in heterogeneous graphs. Extensive experiments on benchmark datasets demonstrate the effectiveness of RASH across various downstream tasks. The code is available at: https://github.com/zhengziyu77/RASH.

Discrepancy-Aware Graph Mask Auto-Encoder

Masked Graph Auto-Encoder, a powerful graph self-supervised training paradigm, has recently shown superior performance in graph representation learning. Existing works typically rely on node contextual information to recover the masked information. However, they fail to generalize well to heterophilic graphs where connected nodes may be not similar, because they focus only on capturing the neighborhood information and ignoring the discrepancy information between different nodes, resulting in indistinguishable node representations. In this paper, to address this issue, we propose a Discrepancy-Aware Graph Mask Auto-Encoder (DGMAE). It obtains more distinguishable node representations by reconstructing the discrepancy information of neighboring nodes during the masking process. We conduct extensive experiments on 17 widely-used benchmark datasets. The results show that our DGMAE can effectively preserve the discrepancies of nodes in low-dimensional space. Moreover, DGMAE significantly outperforms state-of-the-art graph self-supervised learning methods on three graph analytic including tasks node classification, node clustering, and graph classification, demonstrating its remarkable superiority. The code of DGMAE is available at https://github.com/zhengziyu77/DGMAE.

Multiple Object Tracking in Video SAR: A Benchmark and Tracking Baseline

In the context of multi-object tracking using video synthetic aperture radar (Video SAR), Doppler shifts induced by target motion result in artifacts that are easily mistaken for shadows caused by static occlusions. Moreover, appearance changes of the target caused by Doppler mismatch may lead to association failures and disrupt trajectory continuity. A major limitation in this field is the lack of public benchmark datasets for standardized algorithm evaluation. To address the above challenges, we collected and annotated 45 video SAR sequences containing moving targets, and named the Video SAR MOT Benchmark (VSMB). Specifically, to mitigate the effects of trailing and defocusing in moving targets, we introduce a line feature enhancement mechanism that emphasizes the positive role of motion shadows and reduces false alarms induced by static occlusions. In addition, to mitigate the adverse effects of target appearance variations, we propose a motion-aware clue discarding mechanism that substantially improves tracking robustness in Video SAR. The proposed model achieves state-of-the-art performance on the VSMB, and the dataset and model are released at https://github.com/softwarePupil/VSMB.

J-DDL: Surface Damage Detection and Localization System for Fighter Aircraft

Ensuring the safety and extended operational life of fighter aircraft necessitates frequent and exhaustive inspections. While surface defect detection is feasible for human inspectors, manual methods face critical limitations in scalability, efficiency, and consistency due to the vast surface area, structural complexity, and operational demands of aircraft maintenance. We propose a smart surface damage detection and localization system for fighter aircraft, termed J-DDL. J-DDL integrates 2D images and 3D point clouds of the entire aircraft surface, captured using a combined system of laser scanners and cameras, to achieve precise damage detection and localization. Central to our system is a novel damage detection network built on the YOLO architecture, specifically optimized for identifying surface defects in 2D aircraft images. Key innovations include lightweight Fasternet blocks for efficient feature extraction, an optimized neck architecture incorporating Efficient Multiscale Attention (EMA) modules for superior feature aggregation, and the introduction of a novel loss function, Inner-CIOU, to enhance detection accuracy. After detecting damage in 2D images, the system maps the identified anomalies onto corresponding 3D point clouds, enabling accurate 3D localization of defects across the aircraft surface. Our J-DDL not only streamlines the inspection process but also ensures more comprehensive and detailed coverage of large and complex aircraft exteriors. To facilitate further advancements in this domain, we have developed the first publicly available dataset specifically focused on aircraft damage. Experimental evaluations validate the effectiveness of our framework, underscoring its potential to significantly advance automated aircraft inspection technologies.

HF-VTON: High-Fidelity Virtual Try-On via Consistent Geometric and Semantic Alignment

Virtual try-on technology has become increasingly important in the fashion and retail industries, enabling the generation of high-fidelity garment images that adapt seamlessly to target human models. While existing methods have achieved notable progress, they still face significant challenges in maintaining consistency across different poses. Specifically, geometric distortions lead to a lack of spatial consistency, mismatches in garment structure and texture across poses result in semantic inconsistency, and the loss or distortion of fine-grained details diminishes visual fidelity. To address these challenges, we propose HF-VTON, a novel framework that ensures high-fidelity virtual try-on performance across diverse poses. HF-VTON consists of three key modules: (1) the Appearance-Preserving Warp Alignment Module (APWAM), which aligns garments to human poses, addressing geometric deformations and ensuring spatial consistency; (2) the Semantic Representation and Comprehension Module (SRCM), which captures fine-grained garment attributes and multi-pose data to enhance semantic representation, maintaining structural, textural, and pattern consistency; and (3) the Multimodal Prior-Guided Appearance Generation Module (MPAGM), which integrates multimodal features and prior knowledge from pre-trained models to optimize appearance generation, ensuring both semantic and geometric consistency. Additionally, to overcome data limitations in existing benchmarks, we introduce the SAMP-VTONS dataset, featuring multi-pose pairs and rich textual annotations for a more comprehensive evaluation. Experimental results demonstrate that HF-VTON outperforms state-of-the-art methods on both VITON-HD and SAMP-VTONS, excelling in visual fidelity, semantic consistency, and detail preservation.

Conf-GNNRec: Quantifying and Calibrating the Prediction Confidence for GNN-based Recommendation Methods

Recommender systems based on graph neural networks perform well in tasks such as rating and ranking. However, in real-world recommendation scenarios, noise such as user misuse and malicious advertisement gradually accumulates through the message propagation mechanism. Even if existing studies mitigate their effects by reducing the noise propagation weights, the severe sparsity of the recommender system still leads to the low-weighted noisy neighbors being mistaken as meaningful information, and the prediction result obtained based on the polluted nodes is not entirely trustworthy. Therefore, it is crucial to measure the confidence of the prediction results in this highly noisy framework. Furthermore, our evaluation of the existing representative GNN-based recommendation shows that it suffers from overconfidence. Based on the above considerations, we propose a new method to quantify and calibrate the prediction confidence of GNN-based recommendations (Conf-GNNRec). Specifically, we propose a rating calibration method that dynamically adjusts excessive ratings to mitigate overconfidence based on user personalization. We also design a confidence loss function to reduce the overconfidence of negative samples and effectively improve recommendation performance. Experiments on public datasets demonstrate the validity of Conf-GNNRec in prediction confidence and recommendation performance.

Date Fragments: A Hidden Bottleneck of Tokenization for Temporal Reasoning

Modern BPE tokenizers often split calendar dates into meaningless fragments, e.g., 20250312 $\rightarrow$ 202, 503, 12, inflating token counts and obscuring the inherent structure needed for robust temporal reasoning. In this work, we (1) introduce a simple yet interpretable metric, termed date fragmentation ratio, that measures how faithfully a tokenizer preserves multi-digit date components; (2) release DateAugBench, a suite of 6500 examples spanning three temporal reasoning tasks: context-based date resolution, format-invariance puzzles, and date arithmetic across historical, contemporary, and future regimes; and (3) through layer-wise probing and causal attention-hop analyses, uncover an emergent date-abstraction mechanism whereby large language models stitch together the fragments of month, day, and year components for temporal reasoning. Our experiments show that excessive fragmentation correlates with accuracy drops of up to 10 points on uncommon dates like historical and futuristic dates. Further, we find that the larger the model, the faster the emergent date abstraction that heals date fragments is accomplished. Lastly, we observe a reasoning path that LLMs follow to assemble date fragments, typically differing from human interpretation (year $\rightarrow$ month $\rightarrow$ day).

Selftok: Discrete Visual Tokens of Autoregression, by Diffusion, and for Reasoning

We completely discard the conventional spatial prior in image representation and introduce a novel discrete visual tokenizer: Self-consistency Tokenizer (Selftok). At its design core, we compose an autoregressive (AR) prior -- mirroring the causal structure of language -- into visual tokens by using the reverse diffusion process of image generation. The AR property makes Selftok fundamentally distinct from traditional spatial tokens in the following two key ways: - Selftok offers an elegant and minimalist approach to unify diffusion and AR for vision-language models (VLMs): By representing images with Selftok tokens, we can train a VLM using a purely discrete autoregressive architecture -- like that in LLMs -- without requiring additional modules or training objectives. - We theoretically show that the AR prior satisfies the Bellman equation, whereas the spatial prior does not. Therefore, Selftok supports reinforcement learning (RL) for visual generation with effectiveness comparable to that achieved in LLMs. Besides the AR property, Selftok is also a SoTA tokenizer that achieves a favorable trade-off between high-quality reconstruction and compression rate. We use Selftok to build a pure AR VLM for both visual comprehension and generation tasks. Impressively, without using any text-image training pairs, a simple policy gradient RL working in the visual tokens can significantly boost the visual generation benchmark, surpassing all the existing models by a large margin. Therefore, we believe that Selftok effectively addresses the long-standing challenge that visual tokens cannot support effective RL. When combined with the well-established strengths of RL in LLMs, this brings us one step closer to realizing a truly multimodal LLM. Project Page: https://selftok-team.github.io/report/.