arXiv subjects

Chenyang Yu

Publications and source records attributed to Chenyang Yu.

At least 19 recordsLinked to original sources

arXiv2026

Leaving, Staying, or Living There? Duration Commonsense Reasoning with Human-LLM Co-annotations

Gardening in the backyard implies living there, while checking into an Airbnb more likely means staying for a few days. Such duration judgments are important for understanding human mobility, travel behavior, and social context. However, duration is rarely stated explicitly in social media posts and must instead be inferred from noisy textual, visual, and contextual cues. In this paper, we investigate whether vision-language models possess such duration commonsense. We create a new corpus of geotagged tweets co-annotated by humans and LLMs to serve as a testbed for this task. We find that state-of-the-art models struggle with duration inference, even when guided by multi-turn Chain-of-Thought prompting. Error analyses further reveal that models often fail to integrate critical contextual information, or they overly rely on prominent visual details while overlooking subtle hashtags and semantic nuances. These findings underscore the significant challenges remaining in context-aware multimodal understanding.

cs.AI↗

arXiv2026

STEP: Soft Trajectory Estimation of Prediction Correctness via Per-Step Probability Scoring

Estimating the correctness of large language model (LLM) predictions without access to ground truth is critical for reliable deployment in high-stakes settings. Existing approaches, such as entropy and perplexity, operate solely on final output statistics, discarding the rich information encoded in intermediate reasoning steps. To address this limitation, we propose STEP (Soft Trajectory Estimation of Prediction correctness), a training-free framework that scores each reasoning step via lightweight forward passes to obtain a soft probability distribution qi over candidate answers, from which we derive a set of soft trajectory features capturing per-step confidence and its progression toward the final answer. Extensive experiments on various benchmarks demonstrate consistent and substantial improvements, particularly in AUROC, over output-based baselines and hard-indicator trajectory methods. Notably, Temporally-Weighted Consistency, which assigns higher importance to later reasoning steps where answer distributions have stabilized, contributes the largest gains, highlighting the importance of temporal structure in reasoning trajectories. These results further validate the advantage of preserving continuous probability signals over binary discretization.

cs.CL↗

arXiv2026

Streaming-WAM: Action-Conditioned World-Action Model for Asynchronous Robot Manipulation

World action models (WAMs) that use future visual prediction at inference time incur substantial generation costs. Asynchronous execution reduces waiting by overlapping inference with robot motion, but visual predictions used for subsequent action generation must anticipate the effects of actions already scheduled for execution during inference. We introduce Streaming-WAM, which couples action-conditioned world modeling with asynchronous robot control to account for committed actions in future visual prediction. At each streaming update, the model conditions future visual prediction on the latest observation and the committed actions, which form the fixed prefix of the next action chunk. The resulting action-conditioned visual features guide generation of the remaining actions within the same joint update, so the continuation is informed by the scene changes expected during execution of the fixed prefix. On LIBERO, Streaming-WAM achieves an average success rate of 98.35\% and reduces mean episode time by a factor of 2.93 relative to Fast-WAM. On the real-world Stamp Paper task, mean episode time falls from 90 s with synchronous Joint-WAM to 38 s with Streaming-WAM. These results show that Streaming-WAM supports efficient asynchronous control while maintaining high task success rates.

cs.RO↗

arXiv2026

HAT-4D: Lifting Monocular Video for 4D Multi-Object Interactions via Human-Agent Collaboration

Extracting dynamic 4D object interactions from massive, in-the-wild monocular videos offers a highly efficient data collection pathway for scaling Embodied AI and training VLAs. However, existing monocular 4D reconstruction methods primarily focus on isolated objects, often failing under the severe occlusions and complex dynamics inherent in multi-object interactions. To bridge this gap, we propose HAT-4D, the first agentic framework designed to reconstruct the 3D geometry, temporal dynamics, and physical interactions of multiple objects from a single video. By integrating VLMs with a multi-level human-in-the-loop feedback mechanism, HAT-4D efficiently resolves depth ambiguities and interaction-induced occlusions during 3D generation and 4D propagation, yielding physically plausible assets without relying on expensive multicamera rigs. As a scalable data engine, HAT-4D facilitates the creation of MVOIK-4D, an open-world benchmark for monocular 4D interaction reconstruction, accompanied by a novel multi-dimensional evaluation protocol focused on physical plausibility and temporal consistency. Extensive experiments demonstrate that HAT-4D achieves SOTA performance on most evaluation metrics, while maintaining competitive semantic alignment. Ablation studies show that introducing a small amount of human feedback improves interaction reconstruction. Moreover, the data produced by HAT-4D effectively improves baseline performance when used for fine-tuning. Our data and code are available at https://lijiaxin0111.github.io/HAT4D/

cs.CV↗

arXiv2026

New Results Towards the Characterization of Service Rate Region of Reed-Muller Codes

The Service Rate Region (SRR) serves as a critical metric for evaluating the concurrent service capacity of distributed storage systems. While several works have characterized the SRR for MDS codes and first order Reed-Muller codes, for high-order Reed-Muller codes the problem becomes way more complicated and only partial results were given by Ly, Soljanin, and Lalitha [IEEE ISIT 2025]. In this paper, we refine the SRR analysis by explicitly characterizing the intersection patterns of recovery sets for high-order Reed-Muller codes, deriving the exact region for the case m=r+1 and providing new types of strictly tighter constraints to bridge the gap between existing approximations and the exact SRR polytope.

cs.IT↗

arXiv2026

Abstention as an Action Can Kill Both the Reward Gradient and the KL Anchor: Collapse Law and Repair for Error-Penalized Reinforcement Learning

Error-penalized scoring rules ($+1$ for a correct answer, $-λ$ for a wrong one, $0$ for abstaining) are increasingly prescribed against hallucination: a rational agent facing such a rule answers exactly when its correctness probability exceeds Chow's threshold $t^\ast=λ/(1+λ)$. We prove that a KL-anchored gradient learner can do the opposite. When abstention is a discrete action, the reward gradient and the anchor's restoring force are throttled by the same gate-saturation factor and die together: under explicit conditions (among them, blanket answering loses score in expectation and prompts share a bounded readout) the model drifts toward refusing everything, its mean training reward rising to zero like $1/t$ in training time $t$, so the curve reads as improvement while coverage collapses. The advantage estimator compounds the failure: in its sparse-answer regime, group normalization silently replaces every designed penalty with an effective penalty of one, moving the learned threshold from $t^\ast$ to $1/2$. The repair is structural: train a mandatory confidence report with a strictly proper score plus a correctness reward, and abstain only at deployment by thresholding the report. The always-emitted report has no gate to saturate, so no shared factor can kill its reward gradient and its anchor together, and its calibrated optimum is attracting. Simulations confirm every prediction, and experiments on language models at two scales confirm the mechanism live: the rule silences questions the models demonstrably still solve within ten optimizer steps, an ablation isolates the cause, and report-level training raises coverage, accuracy, and calibration together.

cs.LG↗

arXiv2026

QEncodeBench: Can Large Language Models Encode Classical Problems into Verified Quantum Oracles?

Grover search, amplitude amplification, and quantum counting all rely on the same reusable subroutine, a phase oracle, whose construction the algorithms literature takes as given: the classical predicate is assumed to be already encoded as a correct, resource-bounded circuit. We turn this assumption into a measured capability. QEncodeBench tasks large language models (LLMs) with encoding classical constraint problems as phase oracles and scores the generated circuits with an adversarially self-validated verifier that decides full solution-set equivalence up to a global phase, with ancillas restored and resource budgets enforced. Sampled basis-state tests, we show, systematically overestimate this ability. Measured this way, models separate sharply: code models without a reasoning mode solve essentially nothing, and enabling native reasoning on identical weights improves accuracy by an order of magnitude. The failures are overwhelmingly semantic rather than syntactic. Two architectures, a unit-verified constraint agent and a neuro-symbolic compilation pipeline, close most of the remaining gap by delegating correctness-critical composition to deterministic procedures. Ablations quantify the contribution of each component, and resource gating exposes an architecture-dependent trade-off between circuit width and depth. Finally, controlled difficulty escalation reveals architecture-specific responses to difficulty structure: different difficulty axes degrade different methods, while the neuro-symbolic pipeline passes every evaluated instance. Code and data are available at https://github.com/chexujun/QEncodeBench.

quant-ph↗

arXiv2026

AetheRock: An Arm-Worn Robot Teaching System for Force-Guided Vision-Tactile Learning

Force and tactile sensing are indispensable in contact-rich manipulation. However, force-aware robot learning faces critical challenges due to the incompatible assembly of tactile and force sensors in handheld or wearable devices. To address these limitations, we first introduce AetheRock for gripper-force, vision, and tactile data collection, which is an arm-worn device featuring a modular and easily manufactured visuo-tactile sensor, GelSlim-MiniFab, at the fingertip, a resistive pressure sensor at the human finger contact region, a customized PCB module, and a wearable kit for comfortable and robust collection. Building on this, we propose ForceVT, a representation learning framework that uses force and vision to guide fidelity-agnostic tactile learning, enabling robust inference in any tactile situation. Real-world experiments show that AetheRock achieves qualified data efficiency and that ForceVT effectively alleviates inefficiencies when visuo-tactile sensors exhibit manufacturing and utilization inconsistencies. Overall, our work mitigates the limitations of gripper-force vision-tactile robot learning through innovative hardware design and algorithms.

cs.RO↗

arXiv2026

Power Battery Detection

Power batteries are essential components in electric vehicles, where internal structural defects can pose serious safety risks. We conduct a comprehensive study on a new task, power battery detection (PBD), which aims to localize the dense endpoints of cathode and anode plates from industrial X-ray images for quality inspection. Manual inspection is inefficient and error-prone, while traditional vision algorithms struggle with densely packed plates, low contrast, scale variation, and imaging artifacts. To address this issue and drive more attention into this meaningful task, we present PBD5K, the first large-scale benchmark for this task, consisting of 5,000 X-ray images from nine battery types with fine-grained annotations and eight types of real-world visual interference. To support scalable and consistent labeling, we develop an intelligent annotation pipeline that combines image filtering, model-assisted pre-labeling, cross-verification, and layered quality evaluation. We formulate PBD as a point-level segmentation problem and propose MDCNeXt, a model designed to extract and integrate multi-dimensional structure clues including point, line, and count information from the plate itself. To improve discrimination between plates and suppress visual interference, MDCNeXt incorporates two state space modules. The first is a prompt-filtered module that learns contrastive relationships guided by task-specific prompts. The second is a density-aware reordering module that refines segmentation in regions with high plate density. In addition, we propose a distance-adaptive mask generation strategy to provide robust supervision under varying spatial distributions of anode and cathode positions. The source code and datasets will be publicly available at \href{https://github.com/Xiaoqi-Zhao-DLUT/X-ray-PBD}{PBD5K}.

cs.CV↗

arXiv2025

X-ReID: Multi-granularity Information Interaction for Video-Based Visible-Infrared Person Re-Identification

Large-scale vision-language models (e.g., CLIP) have recently achieved remarkable performance in retrieval tasks, yet their potential for Video-based Visible-Infrared Person Re-Identification (VVI-ReID) remains largely unexplored. The primary challenges are narrowing the modality gap and leveraging spatiotemporal information in video sequences. To address the above issues, in this paper, we propose a novel cross-modality feature learning framework named X-ReID for VVI-ReID. Specifically, we first propose a Cross-modality Prototype Collaboration (CPC) to align and integrate features from different modalities, guiding the network to reduce the modality discrepancy. Then, a Multi-granularity Information Interaction (MII) is designed, incorporating short-term interactions from adjacent frames, long-term cross-frame information fusion, and cross-modality feature alignment to enhance temporal modeling and further reduce modality gaps. Finally, by integrating multi-granularity information, a robust sequence-level representation is achieved. Extensive experiments on two large-scale VVI-ReID benchmarks (i.e., HITSZ-VCM and BUPTCampus) demonstrate the superiority of our method over state-of-the-art methods. The source code is released at https://github.com/AsuradaYuci/X-ReID.

cs.CV↗

arXiv2025

FUSE-Traffic: Fusion of Unstructured and Structured Data for Event-aware Traffic Forecasting

Accurate traffic forecasting is a core technology for building Intelligent Transportation Systems (ITS), enabling better urban resource allocation and improved travel experiences. With growing urbanization, traffic congestion has intensified, highlighting the need for reliable and responsive forecasting models. In recent years, deep learning, particularly Graph Neural Networks (GNNs), has emerged as the mainstream paradigm in traffic forecasting. GNNs can effectively capture complex spatial dependencies in road network topology and dynamic temporal evolution patterns in traffic flow data. Foundational models such as STGCN and GraphWaveNet, along with more recent developments including STWave and D2STGNN, have achieved impressive performance on standard traffic datasets. These approaches incorporate sophisticated graph convolutional structures and temporal modeling mechanisms, demonstrating particular effectiveness in capturing and forecasting traffic patterns characterized by periodic regularities. To address this challenge, researchers have explored various ways to incorporate event information. Early attempts primarily relied on manually engineered event features. For instance, some approaches introduced manually defined incident effect scores or constructed specific subgraphs for different event-induced traffic conditions. While these methods somewhat enhance responsiveness to specific events, their core drawback lies in a heavy reliance on domain experts' prior knowledge, making generalization to diverse and complex unknown events difficult, and low-dimensional manual features often lead to the loss of rich semantic details.

cs.LG↗

arXiv2025

UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation

Multi-modal image segmentation faces real-world deployment challenges from incomplete/corrupted modalities degrading performance. While existing methods address training-inference modality gaps via specialized per-combination models, they introduce high deployment costs by requiring exhaustive model subsets and model-modality matching. In this work, we propose a unified modality-relax segmentation network (UniMRSeg) through hierarchical self-supervised compensation (HSSC). Our approach hierarchically bridges representation gaps between complete and incomplete modalities across input, feature and output levels. % First, we adopt modality reconstruction with the hybrid shuffled-masking augmentation, encouraging the model to learn the intrinsic modality characteristics and generate meaningful representations for missing modalities through cross-modal fusion. % Next, modality-invariant contrastive learning implicitly compensates the feature space distance among incomplete-complete modality pairs. Furthermore, the proposed lightweight reverse attention adapter explicitly compensates for the weak perceptual semantics in the frozen encoder. Last, UniMRSeg is fine-tuned under the hybrid consistency constraint to ensure stable prediction under all modality combinations without large performance fluctuations. Without bells and whistles, UniMRSeg significantly outperforms the state-of-the-art methods under diverse missing modality scenarios on MRI-based brain tumor segmentation, RGB-D semantic segmentation, RGB-D/T salient object segmentation. The code will be released at https://github.com/Xiaoqi-Zhao-DLUT/UniMRSeg.

cs.CV↗

arXiv2025

A Decade of Metric Differential Privacy: Advancements and Applications

Metric Differential Privacy (mDP) builds upon the core principles of Differential Privacy (DP) by incorporating various distance metrics, which offer adaptable and context-sensitive privacy guarantees for a wide range of applications, such as location-based services, text analysis, and image processing. Since its inception in 2013, mDP has garnered substantial research attention, advancing theoretical foundations, algorithm design, and practical implementations. Despite this progress, existing surveys mainly focus on traditional DP and local DP, and they provide limited coverage of mDP. This paper provides a comprehensive survey of mDP research from 2013 to 2024, tracing its development from the foundations of DP. We categorize essential mechanisms, including Laplace, Exponential, and optimization-based approaches, and assess their strengths, limitations, and application domains. Additionally, we highlight key challenges and outline future research directions to encourage innovation and real-world adoption of mDP. This survey is designed to be a valuable resource for researchers and practitioners aiming to deepen their understanding and drive progress in mDP within the broader privacy ecosystem.

cs.CR↗

arXiv2025

Inspiring the Next Generation of Segment Anything Models: Comprehensively Evaluate SAM and SAM 2 with Diverse Prompts Towards Context-Dependent Concepts under Different Scenes

As large-scale foundation models trained on billions of image--mask pairs covering a vast diversity of scenes, objects, and contexts, SAM and its upgraded version, SAM~2, have significantly influenced multiple fields within computer vision. Leveraging such unprecedented data diversity, they exhibit strong open-world segmentation capabilities, with SAM~2 further enhancing these capabilities to support high-quality video segmentation. While SAMs (SAM and SAM~2) have demonstrated excellent performance in segmenting context-independent concepts like people, cars, and roads, they overlook more challenging context-dependent (CD) concepts, such as visual saliency, camouflage, industrial defects, and medical lesions. CD concepts rely heavily on global and local contextual information, making them susceptible to shifts in different contexts, which requires strong discriminative capabilities from the model. The lack of comprehensive evaluation of SAMs limits understanding of their performance boundaries, which may hinder the design of future models. In this paper, we conduct a thorough evaluation of SAMs on 11 CD concepts across 2D and 3D images and videos in various visual modalities within natural, medical, and industrial scenes. We develop a unified evaluation framework for SAM and SAM~2 that supports manual, automatic, and intermediate self-prompting, aided by our specific prompt generation and interaction strategies. We further explore the potential of SAM~2 for in-context learning and introduce prompt robustness testing to simulate real-world imperfect prompts. Finally, we analyze the benefits and limitations of SAMs in understanding CD concepts and discuss their future development in segmentation tasks.

cs.CV↗

arXiv2025

PA-Boot: A Formally Verified Authentication Protocol for Multiprocessor Secure Boot

Hardware supply-chain attacks are raising significant security threats to the boot process of multiprocessor systems. This paper identifies a new, prevalent hardware supply-chain attack surface that can bypass multiprocessor secure boot due to the absence of processor-authentication mechanisms. To defend against such attacks, we present PA-Boot, the first formally verified processor-authentication protocol for secure boot in multiprocessor systems. PA-Boot is proved functionally correct and is guaranteed to detect multiple adversarial behaviors, e.g., processor replacements, man-in-the-middle attacks, and tampering with certificates. The fine-grained formalization of PA-Boot and its fully mechanized security proofs are carried out in the Isabelle/HOL theorem prover with 306 lemmas/theorems and ~7,100 LoC. Experiments on a proof-of-concept implementation indicate that PA-Boot can effectively identify boot-process attacks with a considerably minor overhead and thereby improve the security of multiprocessor systems.

cs.CR↗

arXiv2025

FlowDistill: Scalable Traffic Flow Prediction via Distillation from LLMs

Accurate traffic flow prediction is vital for optimizing urban mobility, yet it remains difficult in many cities due to complex spatio-temporal dependencies and limited high-quality data. While deep graph-based models demonstrate strong predictive power, their performance often comes at the cost of high computational overhead and substantial training data requirements, making them impractical for deployment in resource-constrained or data-scarce environments. We propose the FlowDistill, a lightweight and scalable traffic prediction framework based on knowledge distillation from large language models (LLMs). In this teacher-student setup, a fine-tuned LLM guides a compact multi-layer perceptron (MLP) student model using a novel combination of the information bottleneck principle and teacher-bounded regression loss, ensuring the distilled model retains only essential and transferable knowledge. Spatial and temporal correlations are explicitly encoded to enhance the model's generalization across diverse urban settings. Despite its simplicity, FlowDistill consistently outperforms state-of-the-art models in prediction accuracy while requiring significantly less training data, and achieving lower memory usage and inference latency, highlighting its efficiency and suitability for real-world, scalable deployment.

cs.LG↗

arXiv2024

Protecting Vehicle Location Privacy with Contextually-Driven Synthetic Location Generation

Geo-obfuscation is a Location Privacy Protection Mechanism used in location-based services that allows users to report obfuscated locations instead of exact ones. A formal privacy criterion, geoindistinguishability (Geo-Ind), requires real locations to be hard to distinguish from nearby locations (by attackers) based on their obfuscated representations. However, Geo-Ind often fails to consider context, such as road networks and vehicle traffic conditions, making it less effective in protecting the location privacy of vehicles, of which the mobility are heavily influenced by these factors. In this paper, we introduce VehiTrack, a new threat model to demonstrate the vulnerability of Geo-Ind in protecting vehicle location privacy from context-aware inference attacks. Our experiments demonstrate that VehiTrack can accurately determine exact vehicle locations from obfuscated data, reducing average inference errors by 61.20% with Laplacian noise and 47.35% with linear programming (LP) compared to traditional Bayesian attacks. By using contextual data like road networks and traffic flow, VehiTrack effectively eliminates a significant number of seemingly "impossible" locations during its search for the actual location of the vehicles. Based on these insights, we propose TransProtect, a new geo-obfuscation approach that limits obfuscation to realistic vehicle movement patterns, complicating attackers' ability to differentiate obfuscated from actual locations. Our results show that TransProtect increases VehiTrack's inference error by 57.75% with Laplacian noise and 27.21% with LP, significantly enhancing protection against these attacks.

cs.CR↗

arXiv2024

Harnessing LLMs for Cross-City OD Flow Prediction

Understanding and predicting Origin-Destination (OD) flows is crucial for urban planning and transportation management. Traditional OD prediction models, while effective within single cities, often face limitations when applied across different cities due to varied traffic conditions, urban layouts, and socio-economic factors. In this paper, by employing Large Language Models (LLMs), we introduce a new method for cross-city OD flow prediction. Our approach leverages the advanced semantic understanding and contextual learning capabilities of LLMs to bridge the gap between cities with different characteristics, providing a robust and adaptable solution for accurate OD flow prediction that can be transferred from one city to another. Our novel framework involves four major components: collecting OD training datasets from a source city, instruction-tuning the LLMs, predicting destination POIs in a target city, and identifying the locations that best match the predicted destination POIs. We introduce a new loss function that integrates POI semantics and trip distance during training. By extracting high-quality semantic features from human mobility and POI data, the model understands spatial and functional relationships within urban spaces and captures interactions between individuals and various POIs. Extensive experimental results demonstrate the superiority of our approach over the state-of-the-art learning-based methods in cross-city OD flow prediction.

cs.AI↗