arXiv subjects

Jiale Zhang

Publications and source records attributed to Jiale Zhang.

At least 19 recordsLinked to original sources

arXiv2026

In-Context VLA: Endowing Vision-Language-Action Models with Language via In-Context Post-Training and Agentic Tool Use

Vision-Language-Action (VLA) models have become the dominant recipe for generalist manipulation, yet they are almost universally trained by behavior cloning: a policy imitates expert action chunks conditioned on a static image and a fixed instruction. A natural remedy is to inject explicit reasoning through textual chain-of-thought (CoT). We show, both empirically and analytically, that free-form textual CoT degrades low-level control: the reasoning it produces is ungrounded, its latency breaks closed-loop timing, and, crucially, the reasoning and action tokens are optimized against conflicting objectives so that the policy learns to narrate rather than to act. We argue that what a VLA needs is not the ability to generate language, but the ability to consume grounded language. To this end we introduce \textbf{\ourmethod{}}, a framework that endows a VLA with language competence through (i) in-context post-training, in which perceptual evidence is injected as structured context and the model is supervised only on actions, and (ii) an agentic tool-use interface, in which the policy queries open-vocabulary detectors, monocular depth, and a vision--language model to actively acquire task-relevant information. Rather than emitting a single templated caption, our data engine produces diverse, paraphrased, and evidence-conditioned spatial descriptions, so that the policy learns to interpret language it has never seen verbatim. Across the RoboCasa-GR1, SimplerEnv, and LIBERO simulation benchmarks, together with 8 real-world robot manipulation tasks, our method consistently achieves SOTA results in both performance and efficiency when compared with CoT-based approaches under matched configurations.

cs.RO↗

arXiv2026

GEIS: A Generation-Evaluation-Improvement Loop of Agent Skills for Long-Form Article Generation

Long-form article generation remains difficult for large language models because it combines long context, long instructions, and long outputs. Existing multi-agent pipelines such as STORM improve information coverage by simulating role-specialized agents, but their capabilities are often entangled in prompts and fixed procedures, making them hard to inspect, reuse, or iteratively improve. This paper presents GEIS (Generation-Evaluation-Improvement loop of agent Skills), a loop of named and declarative skills for Wikipedia-style long-form article generation. Implemented and evaluated in Tasi Harness, GEIS composes skills for article writing, browser-based evidence and image collection, diagram rendering, PDF-aware pairwise evaluation, and rule-level skill improvement. Its core writing skill follows Request, Plan, Draft, Audit, Refine, and Deliver; the pairwise evaluation skill produces structured quality reports; and the improvement skill maps recurrent findings into permanent patches to the writing skill in our 20-topic experiment. We evaluate GEIS on 20 Wikipedia Featured Article topics. Under the same generation backend, GEIS improves over the Tasi Harness default writer by 8.0 points on a 100-point PDF quality rubric and outperforms STORM on the two comparable writing dimensions, structural quality and content quality. In the 20-topic improvement experiment, the patched writing skill raises the average score from 82.90 to 86.95, with 17 out of 20 topics improved and the gain mainly coming from content quality. These results show that long-form generation can be reframed from a fixed workflow into an inspectable, modular, and evaluation-guided improvement loop.

cs.CL↗

arXiv2026

JoyAI-Sim: A Simulation-Enabled Interconversion Toolchain for the Embodied Data Pyramid

Generalist robot policies require trustworthy evaluation and robot-usable training data, but both are difficult to scale with physical robots alone. Real-robot trials and demonstrations remain the most faithful source of deployment signals, yet they are slow, costly, and hard to reproduce. We present JoyAI-Sim, a simulation-enabled interconversion toolchain for human-robot aligned model evaluation and data generation, denoted as Robot $\rightleftharpoons$ Simulation $\rightleftharpoons$ Human. On the one hand, the Robot $\rightarrow$ Simulation $\rightarrow$ Human pathway supports human-robot aligned model evaluation by reconstructing real-robot tabletop organization tasks as calibrated digital twins for scalable evaluation, while using human embodied feedback to inspect and refine the naturalness of simulated motions. On the other hand, the Human $\rightarrow$ Simulation $\rightarrow$ Robot pathway supports human-robot aligned data generation: it lifts ego-centric human demonstrations into simulation, checks them under robot physical constraints, and converts them into robot-centered trajectories, annotations, and visual observations. Together, these pathways use the JoySim simulator as both a scalable evaluation layer and a physical consistency filter for robot data generation. We further package the core reconstruction, simulation, rendering, and realism-augmentation modules as cloud services on JD Cloud, turning the system into a reusable and scalable infrastructure for robot data generation and model evaluation.

cs.RO↗

arXiv2026

Awakening the Hydra: Stabilizing Multi-Concept Backdoor Injection in Text-to-Image Diffusion Models

Text-to-image diffusion models are increasingly developed through open-source reuse and repeated downstream fine-tuning, where reused checkpoints are difficult to verify and thus more susceptible to hidden backdoor behaviors. In such ecosystems, a single pretrained model may be sequentially adapted and redistributed by multiple independent parties, allowing multiple concept-specific trigger-target associations to accumulate in the same model. When these associations coexist, semantic conflicts can be amplified in the shared representation space, leading to cross-concept entanglement and degraded generation quality. Notably, instead of strengthening the attack, such accumulation can destabilize previously injected behaviors and reduce attack reliability. In this work, we systematically investigate backdoor attacks under this interference-prone setting and propose Hydra, a unified framework for robust and controlled multi-concept backdoor injection under cumulative and decentralized reuse. Our core insight is that stable backdoor injection under large-scale multi-concept settings requires explicitly constraining trigger semantics while coordinating cross-task interactions during optimization. Specifically, Hydra performs evolutionary trigger search in the text encoder space to identify triggers that are semantically aligned with their target concepts while remaining stable across other injected concepts. It further combines multi-task fine-tuning with trigger-clean regularization to improve training stability under dense multi-concept injection. Extensive experiments across multiple diffusion backbones under rigorous multi-concept settings show that Hydra maintains effective backdoor activation while preserving clean generation fidelity and image quality. For instance, across 8 attackers and 500 concept pairs, Hydra maintains ~95% ASR and strong clean generation.

cs.CR↗

arXiv2026

FAST Observations of Wave-like Structures in the Radio Dynamic Spectrum of AD Leo

M-dwarf flare stars like AD Leo are laboratories for studying intense magnetic activities. The coherent radio bursts they produce are powerful probes of stellar coronal plasma and magnetic fields. In this study, we present high-resolution observations of AD Leo from the Five-hundred-meter Aperture Spherical radio Telescope (FAST) that reveal wave-like structures in its radio dynamic spectrum. The observations show trains of short-duration, narrowband sub-bursts where the central frequency, frequency drift rate, and flux density are all simultaneously modulated with a period of 1.53 s. Notably, modulation of the central frequency is approximately in-phase with that of the drift rate but roughly in anti-phase with that of the flux density. Furthermore, the amplitude of the frequency modulation grows with an e-fold timescale of 2.4 s. We interpret the observed sinusoidal frequency modulations as a possible signature of a magnetohydrodynamic (MHD) wave in the stellar corona. Our work provides a window into stellar coronal seismology and offers an opportunity to infer the local plasma environment via the MHD wave model.

astro-ph.SR↗

arXiv2026

DSBA: Dynamic Stealthy Backdoor Attack with Collaborative Optimization in Self-Supervised Learning

Self-Supervised Learning (SSL) has emerged as a significant paradigm in representation learning thanks to its ability to learn without extensive labeled data, its strong generalization capabilities, and its potential for privacy preservation. However, recent research reveals that SSL models are also vulnerable to backdoor attacks. Existing backdoor attack methods in the SSL context commonly suffer from issues such as high detectability of triggers, feature entanglement, and pronounced out-of-distribution properties in poisoned samples, all of which compromises attack effectiveness and stealthiness. To that, we propose a Dynamic Stealthy Backdoor Attack (DSBA) backed by a new technique we term Collaborative Optimization. This method decouples the attack process into two collaborative optimization layers: the outer-layer optimization trains a backdoor encoder responsible for global feature space remodeling, aiming to achieve precise backdoor implantation while preserving core functionality; meanwhile, the inner-layer optimization employs a dynamically optimized generator to adaptively produce optimally concealed triggers for individual samples, achieving coordinated concealment across feature space and visual space. We also introduce multiple loss functions to dynamically balance attack performance and stealthiness, in which we employ an adaptive weight scheduling mechanism to enhance training stability. Extensive experiments on various mainstream SSL algorithms and five public datasets demonstrate that: (i) DSBA significantly enhances Attack Success Rate (ASR) and stealthiness while maintaining downstream task accuracy; and (ii) DSBA exhibits superior robustness against existing mainstream defense methods.

cs.CR↗

arXiv2026

BadRSSD: Backdoor Attacks on Regularized Self-Supervised Diffusion Models

Self-supervised diffusion models learn high-quality visual representations via latent space denoising. However, their representation layer poses a distinct threat: unlike traditional attacks targeting generative outputs, its unconstrained latent semantic space allows for stealthy backdoors, permitting malicious control upon triggering. In this paper, we propose BadRSSD, the first backdoor attack targeting the representation layer of self-supervised diffusion models. Specifically, it hijacks the semantic representations of poisoned samples with triggers in Principal Component Analysis (PCA) space toward those of a target image, then controls the denoising trajectory during diffusion by applying coordinated constraints across latent, pixel, and feature distribution spaces to steer the model toward generating the specified target. Additionally, we integrate representation dispersion regularization into the constraint framework to maintain feature space uniformity, significantly enhancing attack stealth. This approach preserves normal model functionality (high utility) while achieving precise target generation upon trigger activation (high specificity). Experiments on multiple benchmark datasets demonstrate that BadRSSD substantially outperforms existing attacks in both FID and MSE metrics, reliably establishing backdoors across different architectures and configurations, and effectively resisting state-of-the-art backdoor defenses.

cs.CR↗

arXiv2026

Meta-FC: Meta-Learning with Feature Consistency for Robust and Generalizable Watermarking

Deep learning-based watermarking has made remarkable progress in recent years. To achieve robustness against various distortions, current methods commonly adopt a training strategy where a \underline{\textbf{s}}ingle \underline{\textbf{r}}andom \underline{\textbf{d}}istortion (SRD) is chosen as the noise layer in each training batch. However, the SRD strategy treats distortions independently within each batch, neglecting the inherent relationships among different types of distortions and causing optimization conflicts across batches. As a result, the robustness and generalizability of the watermarking model are limited. To address this issue, we propose a novel training strategy that enhances robustness and generalization via \underline{\textbf{meta}}-learning with \underline{\textbf{f}}eature \underline{\textbf{c}}onsistency (Meta-FC). Specifically, we randomly sample multiple distortions from the noise pool to construct a meta-training task, while holding out one distortion as a simulated ``unknown'' distortion for the meta-testing phase. Through meta-learning, the model is encouraged to identify and utilize neurons that exhibit stable activations across different types of distortions, mitigating the optimization conflicts caused by the random sampling of diverse distortions in each batch. To further promote the transformation of stable activations into distortion-invariant representations, we introduce a feature consistency loss that constrains the decoded features of the same image subjected to different distortions to remain consistent. Extensive experiments demonstrate that, compared to the SRD training strategy, Meta-FC improves the robustness and generalization of various watermarking models by an average of 1.59\%, 4.71\%, and 2.38\% under high-intensity, combined, and unknown distortions.

cs.CV↗

arXiv2026

FireRed-Image-Edit-1.0 Technical Report

We present FireRed-Image-Edit, a diffusion transformer for instruction-based image editing that achieves state-of-the-art performance through systematic optimization of data curation, training methodology, and evaluation design. We construct a 1.6B-sample training corpus, comprising 900M text-to-image and 700M image editing pairs from diverse sources. After rigorous cleaning, stratification, auto-labeling, and two-stage filtering, we retain over 100M high-quality samples balanced between generation and editing, ensuring strong semantic coverage and instruction alignment. Our multi-stage training pipeline progressively builds editing capability via pre-training, supervised fine-tuning, and reinforcement learning. To improve data efficiency, we introduce a Multi-Condition Aware Bucket Sampler for variable-resolution batching and Stochastic Instruction Alignment with dynamic prompt re-indexing. To stabilize optimization and enhance controllability, we propose Asymmetric Gradient Optimization for DPO, DiffusionNFT with layout-aware OCR rewards for text editing, and a differentiable Consistency Loss for identity preservation. We further establish REDEdit-Bench, a comprehensive benchmark spanning 15 editing categories, including newly introduced beautification and low-level enhancement tasks. Extensive experiments on REDEdit-Bench and public benchmarks (ImgEdit and GEdit) demonstrate competitive or superior performance against both open-source and proprietary systems. To support future research, our code, models, and benchmark suite are publicly available at https://github.com/FireRedTeam/FireRed-Image-Edit/ .

cs.CV↗

arXiv2026

ADCA: Attention-Driven Multi-Party Collusion Attack in Federated Self-Supervised Learning

Federated Self-Supervised Learning (FSSL) integrates the privacy advantages of distributed training with the capability of self-supervised learning to leverage unlabeled data, showing strong potential across applications. However, recent studies have shown that FSSL is also vulnerable to backdoor attacks. Existing attacks are limited by their trigger design, which typically employs a global, uniform trigger that is easily detected, gets diluted during aggregation, and lacks robustness in heterogeneous client environments. To address these challenges, we propose the Attention-Driven multi-party Collusion Attack (ADCA). During local pre-training, malicious clients decompose the global trigger to find optimal local patterns. Subsequently, these malicious clients collude to form a malicious coalition and establish a collaborative optimization mechanism within it. In this mechanism, each submits its model updates, and an attention mechanism dynamically aggregates them to explore the best cooperative strategy. The resulting aggregated parameters serve as the initial state for the next round of training within the coalition, thereby effectively mitigating the dilution of backdoor information by benign updates. Experiments on multiple FSSL scenarios and four datasets show that ADCA significantly outperforms existing methods in Attack Success Rate (ASR) and persistence, proving its effectiveness and robustness.

cs.CR↗

arXiv2026

HPE: Hallucinated Positive Entanglement for Backdoor Attacks in Federated Self-Supervised Learning

Federated self-supervised learning (FSSL) enables collaborative training of self-supervised representation models without sharing raw unlabeled data. While it serves as a crucial paradigm for privacy-preserving learning, its security remains vulnerable to backdoor attacks, where malicious clients manipulate local training to inject targeted backdoors. Existing FSSL attack methods, however, often suffer from low utilization of poisoned samples, limited transferability, and weak persistence. To address these limitations, we propose a new backdoor attack method for FSSL, namely Hallucinated Positive Entanglement (HPE). HPE first employs hallucination-based augmentation using synthetic positive samples to enhance the encoder's embedding of backdoor features. It then introduces feature entanglement to enforce tight binding between triggers and backdoor samples in the representation space. Finally, selective parameter poisoning and proximity-aware updates constrain the poisoned model within the vicinity of the global model, enhancing its stability and persistence. Experimental results on several FSSL scenarios and datasets show that HPE significantly outperforms existing backdoor attack methods in performance and exhibits strong robustness under various defense mechanisms.

cs.CR↗

arXiv2025

Grey graphs and its application

In multi-attribute decision-making problems where the attribute values are interval grey numbers, a simplified form based on kernels and the degree of greyness is presented. Combining fuzzy graph theory with the kernel and the degree of greyness of interval grey numbers, grey graphs and their corresponding operation rules are presented. This paper presents a new multi-attribute decision-making method based on grey graph theory. We analyzed and evaluated the alternative schemes using grey graph. Lastly, a numerical example was conducted in order to demonstrate the effectiveness and feasibility of the proposed method.

eess.SY↗

arXiv2025

TagLabel: RFID Based Orientation and Material Sensing for Automated Package Inspection

Modern logistics systems face increasing difficulty in identifying counterfeit products, fraudulent returns, and hazardous items concealed within packages, yet current package screening methods remain too slow, expensive, and impractical for widespread use. This paper presents TagLabel, an RFID based system that determines both the orientation and contents of packages using low cost passive UHF tags. By analyzing how materials change RSSI and phase, the system identifies the contents of a package without opening it. Using orientation inferred from phase differences, tag occlusion, and antenna gain patterns, the system selects the tag with the greatest occlusion for accurate material sensing. We evaluate two and three tag configurations, and show that both can deliver high orientation and material sensing performance through the use of machine learning classifiers, even in realistic RF environments. When combined into a unified pipeline, TagLabel achieves more than 80 percent accuracy across all package orientations. Because it requires only standard RFID hardware and offers fast scanning times, this approach provides a practical way to enhance package inspection and improve automation in logistics operations.

eess.SP↗

arXiv2025

The Evolving Landscape of Interactive Surface Sensing Technologies

Interactive surfaces have evolved from capacitive touch and IR based systems into a diverse ecosystem of sensing technologies that support rich and expressive human computer interaction. This survey traces that progression, beginning with infrared vision based approaches, such as FTIR and diffuse illumination, and the rise of capacitive touch as the dominant technology in modern devices, to focusing on contemporary modalities including vision and acoustic sensing. New technologies under development are also discussed, including mmWave radar, and vibration based techniques. Each sensing technique is examined in terms of its operating principles, resolution, scalability, and applications, along with discussions of multimodal integration. By comparing tradeoffs between sensing modalities, the survey highlights the technical and design factors that shape interactive surface performance and user experience. The review concludes by identifying persistent challenges, including sensing accuracy, power constraints, and privacy concerns, and outlines how emerging sensing modalities can enable future interactive environments to be ubiquitous and intelligent.

eess.SY↗

arXiv2025

Radio Burst Phenomenology of AD Leonis and Associated Signatures of Propagation Effects

We present the high-resolution radio dynamic spectra of AD Leonis (AD Leo) between 1.0 and 1.5 GHz taken by the Five-hundred-meter Aperture Spherical radio Telescope (FAST) on Dec. 1st, 2023. Over a 15-minute period, we identify complex, superimposed spectro-temporal structures, including: (1) broadband, second-long modulation lanes with downward frequency drifts, (2) narrowband ($\approx$ 50 MHz), short-duration S-burst envelopes with upward drifts, and (3) even narrower ($\approx$ 10 MHz), millisecond-scale S-burst striae within these envelopes. Using the discrete Fourier transform and auto-correlation function, we identify two dominant periodic emission patterns, corresponding to the periodicities of the S-bursts ($\approx0.1$ s) and the striae ($\approx0.01$ s). The complex superposition of diverse time-frequency structures poses a challenge to interpreting all the emission variability as intrinsic to the source. We propose that the modulation lanes could be a propagation effect as the radio waves traverse an inhomogeneous, regularly structured plasma region in the AD Leo's magnetosphere. By modelling a plasma screen with sinusoidal phase variation in one dimension, we show that we could qualitatively reconstruct the observed modulation lanes. The origin of the finest structures, the striae, remains unclear. Our work highlights that propagation effects in the stellar magnetosphere can potentially probe kilometre-scale structures in the emission regions and provide novel constraints on density inhomogeneities caused by magnetohydrodynamic waves that are difficult to access by other means.

astro-ph.SR↗

arXiv2025

End-to-end Autonomous Vehicle Following System using Monocular Fisheye Camera

The increase in vehicle ownership has led to increased traffic congestion, more accidents, and higher carbon emissions. Vehicle platooning is a promising solution to address these issues by improving road capacity and reducing fuel consumption. However, existing platooning systems face challenges such as reliance on lane markings and expensive high-precision sensors, which limits their general applicability. To address these issues, we propose a vehicle following framework that expands its capability from restricted scenarios to general scenario applications using only a camera. This is achieved through our newly proposed end-to-end method, which improves overall driving performance. The method incorporates a semantic mask to address causal confusion in multi-frame data fusion. Additionally, we introduce a dynamic sampling mechanism to precisely track the trajectories of preceding vehicles. Extensive closed-loop validation in real-world vehicle experiments demonstrates the system's ability to follow vehicles in various scenarios, outperforming traditional multi-stage algorithms. This makes it a promising solution for cost-effective autonomous vehicle platooning. A complete real-world vehicle experiment is available at https://youtu.be/zL1bcVb9kqQ.

cs.RO↗

arXiv2025

Starspots as the origin of ultrafast drifting radio bursts from an active M dwarf

Detecting coherent radio bursts from nearby M dwarfs provides opportunities for exploring their magnetic activity and interaction with orbiting exoplanets. However, it remains uncertain if the emission is related to flare-like activity similar to the Sun or magnetospheric process akin to magnetized planets. Using observations (1.0 - 1.5 GHz) taken by the Five-hundred-meter Aperture Spherical radio Telescope, we found a type of millisecond-scale radio bursts with exceptionally high frequency drift rates ($\sim 8\;\rm{GHz\;s^{-1}}$) from an active M dwarf, AD Leo. The ultrafast drift rates point to a source region with a notably low magnetic scale height ($<0.15\; r_\star$, $r_\star$ as the stellar radius), a feature not expected in a commonly assumed dipole-like global field but highly possible in localized strong-field structures, i.e. starspots. Our findings suggest that a concentrated magnetic field above starspots could be responsible for some of the most intense radio bursts from M dwarfs, supporting a solar-like electron acceleration mechanism.

astro-ph.SR↗

arXiv2025

Federated Unlearning in the Wild: Rethinking Fairness and Data Discrepancy

Machine unlearning is critical for enforcing data deletion rights like the "right to be forgotten." As a decentralized paradigm, Federated Learning (FL) also requires unlearning, but realistic implementations face two major challenges. First, fairness in Federated Unlearning (FU) is often overlooked. Exact unlearning methods typically force all clients into costly retraining, even those uninvolved. Approximate approaches, using gradient ascent or distillation, make coarse interventions that can unfairly degrade performance for clients with only retained data. Second, most FU evaluations rely on synthetic data assumptions (IID/non-IID) that ignore real-world heterogeneity. These unrealistic benchmarks obscure the true impact of unlearning and limit the applicability of current methods. We first conduct a comprehensive benchmark of existing FU methods under realistic data heterogeneity and fairness conditions. We then propose a novel, fairness-aware FU approach, Federated Cross-Client-Constrains Unlearning (FedCCCU), to explicitly address both challenges. FedCCCU offers a practical and scalable solution for real-world FU. Experimental results show that existing methods perform poorly in realistic settings, while our approach consistently outperforms them.

cs.LG↗