Search arXiv⌕ Search

arXiv subjects

Jian Zhao

Publications and source records attributed to Jian Zhao.

At least 73 records · Page 4Linked to original sources
arXiv2026

When Constraints Limit and Inspire: Characterizing Presentation Authoring Practices for Evolving Narratives

Authoring presentation slides involves navigating contextual constraints that shape how content is structured, adapted, and reused. While prior work frames constraints as limitations, little is known about how presenters actively reason about them. We conducted a formative study with ten presenters to examine how constraints emerge, are interpreted, and influence authoring decisions, leading to the Constraint-based Multi-session Presentation Authoring (CMPA) framework. CMPA treats time, audience, and communicative intent as key constraints shaping authoring. We instantiated CMPA in ReSlide, a research prototype for constraint-aware slide creation and reuse, and conducted two user studies on (1) single-session behaviors and (2) multi-session workflows. Compared to a baseline tool, ReSlide helped presenters treat constraints as active design drivers that guide narrative construction. The second study further shows how presenters flexibly reuse and adapt content across authoring cycles as constraints evolve. We then propose design implications for future constraint-aware presentation tools.

cs.HC↗
arXiv2026

Membership Inference for Contrastive Pre-training Models with Text-only PII Queries

Contrastive pretraining models such as CLIP and CLAP, serve as the ubiquitous perceptual backbones for modern multimodal large models, yet their reliance on web-scale data raises growing concerns about memorizing Personally Identifiable Information (PII). Auditing such models via membership inference is challenging in practice: shadow-model MIAs are computationally prohibitive for large multimodal backbones, and existing multimodal auditing methods typically require querying the target with paired biometric inputs, thereby directly exposing sensitive biometric information to the target model. To bypass this critical limitation, we demonstrate a highly desirable capability for privacy auditing: multimodal memorization within these foundational encoders can be accurately inferred using exclusively the text modality. We propose Unimodal Membership Inference Detector (UMID), a text-only auditing framework that performs text-guided cross-modal latent inversion and extracts two complementary signals, similarity (alignment to the queried text) and variability (consistency across randomized inversions). UMID compares these statistics to a lightweight non-member reference constructed from synthetic gibberish and makes decisions via an ensemble of unsupervised anomaly detectors. Comprehensive experiments across diverse CLIP and CLAP architectures demonstrate that UMID significantly improves the effectiveness and efficiency over prior MIAs, delivering strong detection performance with sub-second auditing cost using solely text queries, completely circumventing the need for biometric inputs and complying with strict privacy constraints.

cs.CR↗
arXiv2026

Freshness-Aware Prioritized Experience Replay for LLM/VLM Reinforcement Learning

Reinforcement Learning (RL) has achieved impressive success in post-training Large Language Models (LLMs) and Vision-Language Models (VLMs), with on-policy algorithms such as PPO, GRPO, and REINFORCE++ serving as the dominant paradigm. However, these methods discard all collected trajectories after a single gradient update, resulting in poor sample efficiency, particularly wasteful for agentic tasks where multi-turn environment interactions are expensive. While Experience Replay drives sample efficiency in classic RL by allowing agents to reuse past trajectories and prioritize informative ones, directly applying Prioritized Experience Replay (PER) to LLMs fails. The rapid policy evolution of billion-parameter models renders stored priorities stale, causing old high-priority trajectories to dominate sampling long after they have become uninformative. We propose Freshness-Aware PER, which addresses this priority staleness problem by augmenting any PER-based priority with a multiplicative exponential age decay grounded in effective sample size analysis. To the best of our knowledge, Freshness-Aware PER is the first work to successfully apply PER to LLM/VLM reinforcement learning. We evaluate on eight multi-step agentic, reasoning, and math competition tasks with 0.5B, 3B, and 7B models. Freshness-Aware PER significantly outperforms on-policy baselines, achieving +46% on NQ Search, +367% on Sokoban, and +133% on VLM FrozenLake, while standard PER without age decay consistently degrades performance. Our code is publicly available at https://github.com/Vision-CAIR/Freshness-Aware-PER.

cs.CL↗
arXiv2026

Pushing the Limits of Distillation-Based Continual Learning via Classifier-Proximal Lightweight Plugins

Continual learning requires models to learn continuously while preserving prior knowledge under evolving data streams. Distillation-based methods are appealing for retaining past knowledge in a shared single-model framework with low storage overhead. However, they remain constrained by the stability-plasticity dilemma: knowledge acquisition and preservation are still optimized through coupled objectives, and existing enhancement methods do not alter this underlying bottleneck. To address this issue, we propose a plugin extension paradigm termed Distillation-aware Lightweight Components (DLC) for distillation-based CL. DLC deploys lightweight residual plugins into the base feature extractor's classifier-proximal layer, enabling semantic-level residual correction for better classification accuracy while minimizing disruption to the overall feature extraction process. During inference, plugin-enhanced representations are aggregated to produce classification predictions. To mitigate interference from non-target plugins, we further introduce a lightweight weighting unit that learns to assign importance scores to different plugin-enhanced representations. DLC could deliver a significant 8% accuracy gain on large-scale benchmarks while introducing only a 4% increase in backbone parameters, highlighting its exceptional efficiency. Moreover, DLC is compatible with other plug-and-play CL enhancements and delivers additional gains when combined with them.

cs.LG↗
arXiv2026

YASA: Scalable Multi-Language Taint Analysis on the Unified AST at Ant Group

Modern enterprises increasingly adopt diverse technology stacks with various programming languages, posing significant challenges for static application security testing (SAST). Existing taint analysis tools are predominantly designed for single languages, requiring substantial engineering effort that scales with language diversity. While multi-language tools like CodeQL, Joern, and WALA attempt to address these challenges, they face limitations in intermediate representation design, analysis precision, and extensibility, which make them difficult to scale effectively for large-scale industrial applications at Ant Group. To bridge this gap, we present YASA (Yet Another Static Analyzer), a unified multi-language static taint analysis framework designed for industrial-scale deployment. Specifically, YASA introduces the Unified Abstract Syntax Tree (UAST) that provides a unified abstraction for compatibility across diverse programming languages. Building on the UAST, YASA performs point-to analysis and taint propagation, leveraging a unified semantic model to manage language-agnostic constructs, while incorporating language-specific semantic models to handle other unique language features. When compared to 6 single- and 2 multi-language static analyzers on an industry-standard benchmark, YASA consistently outperformed all baselines across Java, JavaScript, Python, and Go. In real-world deployment within Ant Group, YASA analyzed over 100 million lines of code across 7.3K internal applications. It identified 314 previously unknown taint paths, with 92 of them confirmed as 0-day vulnerabilities. All vulnerabilities were responsibly reported, with 76 already patched by internal development teams, demonstrating YASA's practical effectiveness for securing large-scale industrial software systems.

cs.SE↗
arXiv2026

SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis

While recent years have witnessed rapid progress in speech synthesis, open-source singing voice synthesis (SVS) systems still face significant barriers to industrial deployment, particularly in terms of robustness and zero-shot generalization. In this report, we introduce SoulX-Singer, a high-quality open-source SVS system designed with practical deployment considerations in mind. SoulX-Singer supports controllable singing generation conditioned on either symbolic musical scores (MIDI) or melodic representations, enabling flexible and expressive control in real-world production workflows. Trained on more than 42,000 hours of vocal data, the system supports Mandarin Chinese, English, and Cantonese and consistently achieves state-of-the-art synthesis quality across languages under diverse musical conditions. Furthermore, to enable reliable evaluation of zero-shot SVS performance in practical scenarios, we construct SoulX-Singer-Eval, a dedicated benchmark with strict training-test disentanglement, facilitating systematic assessment in zero-shot settings.

eess.AS↗
arXiv2026

GameUIAgent: An LLM-Powered Framework for Automated Game UI Design with Structured Intermediate Representation

Game UI design requires consistent visual assets across rarity tiers yet remains a predominantly manual process. We present GameUIAgent, an LLM-powered agentic framework that translates natural language descriptions into editable Figma designs via a Design Spec JSON intermediate representation. A six-stage neuro-symbolic pipeline combines LLM generation, deterministic post-processing, and a Vision-Language Model (VLM)-guided Reflection Controller (RC) for iterative self-correction with guaranteed non-regressive quality. Evaluated across 110 test cases, three LLMs, and three UI templates, cross-model analysis establishes a game-domain failure taxonomy (rarity-dependent degradation; visual emptiness) and uncovers two key empirical findings. A Quality Ceiling Effect (Pearson r=-0.96, p<0.01) suggests that RC improvement is bounded by headroom below a quality threshold -- a visual-domain counterpart to test-time compute scaling laws. A Rendering-Evaluation Fidelity Principle reveals that partial rendering enhancements paradoxically degrade VLM evaluation by amplifying structural defects. Together, these results establish foundational principles for LLM-driven visual generation agents in game production.

cs.AI↗
arXiv2026

CompassNav: Steering From Path Imitation To Decision Understanding In Navigation

The dominant paradigm for training Large Vision-Language Models (LVLMs) in navigation relies on imitating expert trajectories. This approach reduces the complex navigation task to a sequence-to-sequence replication of a single correct path, fundamentally limiting the agent's ability to explore and generalize. In this work, we argue for and introduce a new paradigm: a shift from Path Imitation to Decision Understanding. The goal of this paradigm is to build agents that do not just follow, but truly understand how to navigate. We materialize this through two core contributions: first, we introduce Compass-Data-22k, a novel 22k-trajectory dataset. Its Reinforcement Fine-Tuning (RFT) subset provides a panoramic view of the decision landscape by annotating all feasible actions with A* geodesic distances. Second, we design a novel gap-aware hybrid reward function that dynamically adapts its feedback to decision certainty, shifting between decisive signals for optimal actions and nuanced scores to encourage exploration. Integrated into an SFT-then-RFT recipe, our CompassNav agent is trained not to memorize static routes, but to develop an internal compass that constantly intuits the direction to the goal by evaluating the relative quality of all possible moves. This approach enables our 7B agent to set a new state-of-the-art on Goal navigation benchmarks, outperforming even larger proprietary models, and achieve robust real-world goal navigation on a physical robot.

cs.RO↗
arXiv2026

$π$-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs

Flow-based vision-language-action (VLA) models excel in embodied control but suffer from intractable likelihoods during multi-step sampling, hindering online reinforcement learning. We propose \textbf{\textit{$\boldsymbolπ$-StepNFT}} (Step-wise Negative-aware Fine-Tuning), a critic-and-likelihood-free framework that requires only a single forward pass per optimization step and eliminates auxiliary value networks. We identify that wider exploration spaces necessitate finer-grained, step-wise guidance for alignment. Empirically, $π$-StepNFT unlocks latent potential on LIBERO with competitive few-shot robustness. Moreover, it achieves superior generalization on ManiSkill, outperforming value-based baselines in OOD scenarios by preventing overfitting to multimodal features. This property offers a scalable solution promising for complex real-world applications.

cs.RO↗
arXiv2026

Notational Animating: An Interactive Approach to Creating and Editing Animation Keyframes

We introduce the concept of notational animating, an interaction paradigm for animation authoring where users sketch high-level notations over static drawings to indicate intended motions, which are then interpreted by automatic methods (e.g., GenAI models) to generate animation keyframes. Sketched notations have long served as cognitive instruments for animators, capturing forces, poses, dynamics, paths, and other animation features. However, such notations are often context-dependent, non-categorical, ambiguous, and composable based on our analysis of real-world animator-produced sketches. To facilitate interpretation, we first formalize these notations into a structured animation representation (i.e., source, path, and target). We then built an animation authoring system that translates high-level notations into the formalized intended animation, provides dynamic UI widgets for fine-grained parameter control, and establishes a closed feedback loop to resolve ambiguity. Finally, through a preliminary study with animators, we assess the usability of notational animating, reflect its affordance, and identify its contexts of use.

cs.HC↗
arXiv2026

Challenges in Synchronous & Remote Collaboration Around Visualization

We characterize 16 challenges faced by those investigating and developing remote and synchronous collaborative experiences around visualization. Our work reflects the perspectives and prior research efforts of an international group of 29 experts from across human-computer interaction and visualization sub-communities. The challenges are anchored around five collaborative activities that exhibit a centrality of visualization and multimodal communication. These activities include exploratory data analysis, creative ideation, visualization-rich presentations, joint decision making grounded in data, and real-time data monitoring. The challenges also reflect the changing dynamics of these activities in the face of recent advances in extended reality (XR) and artificial intelligence (AI). As an organizing scheme for future research at the intersection of visualization and computer-supported cooperative work, we align the challenges with a sequence of four sets of research and development activities: technological choices, social factors, AI assistance, and evaluation.

cs.HC↗
arXiv2026

Sim2Sea: Sim-to-Real Policy Transfer for Maritime Vessel Navigation in Congested Waters

Autonomous navigation in congested maritime environments is a critical capability for a wide range of real-world applications. However, it remains an unresolved challenge due to complex vessel interactions and significant environmental uncertainties. Existing methods often fail in practical deployment due to a substantial sim-to-real gap, which stems from imprecise simulation, inadequate situational awareness, and unsafe exploration strategies. To address these, we propose \textbf{Sim2Sea}, a comprehensive framework designed to bridge simulation and real-world execution. Sim2Sea advances in three key aspects. First, we develop a GPU-accelerated parallel simulator for scalable and accurate maritime scenario simulation. Second, we design a dual-stream spatiotemporal policy that handles complex dynamics and multi-modal perception, augmented with a velocity-obstacle-guided action masking mechanism to ensure safe and efficient exploration. Finally, a targeted domain randomization scheme helps bridge the sim-to-real gap. Simulation results demonstrate that our method achieves faster convergence and safer trajectories than established baselines. In addition, our policy trained purely in simulation successfully transfers zero-shot to a 17-ton unmanned vessel operating in real-world congested waters. These results validate the effectiveness of Sim2Sea in achieving reliable sim-to-real transfer for practical autonomous maritime navigation.

cs.RO↗
arXiv2026

Swimming Under Constraints: A Safe Reinforcement Learning Framework for Quadrupedal Bio-Inspired Propulsion

Bio-inspired aquatic propulsion offers high thrust and maneuverability but is prone to destabilizing forces such as lift fluctuations, which are further amplified by six-degree-of-freedom (6-DoF) fluid coupling. We formulate quadrupedal swimming as a constrained optimization problem that maximizes forward thrust while minimizing destabilizing fluctuations. Our proposed framework, Accelerated Constrained Proximal Policy Optimization with a PID-regulated Lagrange multiplier (ACPPO-PID), enforces constraints with a PID-regulated Lagrange multiplier, accelerates learning via conditional asymmetric clipping, and stabilizes updates through cycle-wise geometric aggregation. Initialized with imitation learning and refined through on-hardware towing-tank experiments, ACPPO-PID produces control policies that transfer effectively to quadrupedal free-swimming trials. Results demonstrate improved thrust efficiency, reduced destabilizing forces, and faster convergence compared with state-of-the-art baselines, underscoring the importance of constraint-aware safe RL for robust and generalizable bio-inspired locomotion in complex fluid environments.

cs.RO↗
arXiv2026

The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution

Real-world language agents must handle complex, multi-step workflows across diverse Apps. For instance, an agent may manage emails by coordinating with calendars and file systems, or monitor a production database to detect anomalies and generate reports following an operating manual. However, existing language agent benchmarks often focus on narrow domains or simplified tasks that lack the diversity, realism, and long-horizon complexity required to evaluate agents' real-world performance. To address this gap, we introduce the Tool Decathlon (dubbed as Toolathlon), a benchmark for language agents offering diverse Apps and tools, realistic environment setup, and reliable execution-based evaluation. Toolathlon spans 32 software applications and 604 tools, ranging from everyday platforms such as Google Calendar and Notion to professional ones like WooCommerce, Kubernetes, and BigQuery. Most of the tools are based on a high-quality set of Model Context Protocol (MCP) servers that we may have revised or implemented ourselves. Unlike prior works, which primarily ensure functional realism but offer limited environment state diversity, we provide realistic initial environment states from real software, such as Canvas courses with dozens of students or real financial spreadsheets. This benchmark includes 108 manually sourced or crafted tasks in total, requiring interacting with multiple Apps over around 20 turns on average to complete. Each task is strictly verifiable through dedicated evaluation scripts. Comprehensive evaluation of SOTA models highlights their significant shortcomings: the best-performing model, Claude-4.5-Sonnet, achieves only a 38.6% success rate with 20.2 tool calling turns on average, while the top open-weights model DeepSeek-V3.2-Exp reaches 20.1%. We expect Toolathlon to drive the development of more capable language agents for real-world, long-horizon task execution.

cs.CL↗
arXiv2026

To Slide or Not to Slide: Exploring Techniques for Comparing Immersive Videos

Immersive videos (IVs) provide 360° environments that create a strong sense of presence and spatial exploration. Unlike traditional videos, IVs distribute information across multiple directions, making comparison cognitively demanding and highly dependent on interaction techniques. With the growing adoption of IVs, effective comparison techniques have become an essential yet underexplored area of research. Inspired by the "sliding" concept in 2D media comparison, we integrate two established comparison strategies from the literature--toggle and side-by-side--to support IV comparison with greater flexibility. For an in-depth understanding of different strategies, we adapt and implement five IV comparison techniques across VR and 2D environments: SlideInVR, ToggleInVR, SlideIn2D, ToggleIn2D, and SideBySideIn2D. We then conduct a user study (N=20) to examine how these techniques shape users' perceptions, strategies, and workflows. Our findings provide empirical insights into the strengths and limitations of each technique, underscoring the need to switch between comparison approaches across scenarios. Notably, participants consistently rate SlideInVR and SlideIn2D as the most flexible and favorite methods for IV comparison.

cs.HC↗
arXiv2026

Structure-Level Disentangled Diffusion for Few-Shot Chinese Font Generation

Few-shot Chinese font generation aims to synthesize new characters in a target style using only a handful of reference images. Achieving accurate content rendering and faithful style transfer requires effective disentanglement between content and style. However, existing approaches achieve only feature-level disentanglement, allowing the generator to re-entangle these features, leading to content distortion and degraded style fidelity. We propose the Structure-Level Disentangled Diffusion Model (SLD-Font), which receives content and style information from two separate channels. SimSun-style images are used as content templates and concatenated with noisy latent features as the input. Style features extracted by a CLIP model from target-style images are integrated via cross-attention. Additionally, we train a Background Noise Removal module in the pixel space to remove background noise in complex stroke regions. Based on theoretical validation of disentanglement effectiveness, we introduce a parameter-efficient fine-tuning strategy that updates only the style-related modules. This allows the model to better adapt to new styles while avoiding overfitting to the reference images' content. We further introduce the Grey and OCR metrics to evaluate the content quality of generated characters. Experimental results show that SLD-Font achieves significantly higher style fidelity while maintaining comparable content accuracy to existing state-of-the-art methods.

cs.CV↗
arXiv2026

Evaluating Replay Techniques for Asynchronous Task Handover in Immersive Analytics

Immersive analytics enables collaborative data analysis in shared virtual spaces. While synchronous collaboration in such environments is well-established, real-world analysis often requires an effective task handover - the transfer of knowledge and analytical context between analysts working asynchronously. Traditional handover methods often rely on static annotations that fail to capture the dynamic problem-solving process and spatial context inherent in immersive workflows. To address this handover challenge, we explore session replay as a comprehensive approach for analysts to re-experience a predecessor's work, facilitating a deeper understanding of both the visual details and the insight formation process. Two phases of studies were conducted to establish design guidelines for such replay systems by investigating the impact of viewing platform (PC vs. VR), perspective (first-person vs. third-person), and navigation control (active vs. passive). Phase 1 identified the optimal replay configurations within each viewing platform, revealing a platform-dependent divergence: PC users favored a guided, first-person perspective for its focused detail, while VR users benefited significantly from the agency afforded by a third-person perspective with active navigation. After refining each condition based on user feedback, including developing a novel hybrid 1PP+3PP format for PC, Phase 2 compared the two optimized systems (PC vs. VR). Our results show that the immersive VR replay led to significantly better task comprehension and workflow reconstruction accuracy, demonstrating the critical role of embodied agency in understanding complex analytical processes.

cs.HC↗
arXiv2026

Adaptive Image Zoom-in with Bounding Box Transformation for UAV Object Detection

Detecting objects from UAV-captured images is challenging due to the small object size. In this work, a simple and efficient adaptive zoom-in framework is explored for object detection on UAV images. The main motivation is that the foreground objects are generally smaller and sparser than those in common scene images, which hinders the optimization of effective object detectors. We thus aim to zoom in adaptively on the objects to better capture object features for the detection task. To achieve the goal, two core designs are required: \textcolor{black}{i) How to conduct non-uniform zooming on each image efficiently? ii) How to enable object detection training and inference with the zoomed image space?} Correspondingly, a lightweight offset prediction scheme coupled with a novel box-based zooming objective is introduced to learn non-uniform zooming on the input image. Based on the learned zooming transformation, a corner-aligned bounding box transformation method is proposed. The method warps the ground-truth bounding boxes to the zoomed space to learn object detection, and warps the predicted bounding boxes back to the original space during inference. We conduct extensive experiments on three representative UAV object detection datasets, including VisDrone, UAVDT, and SeaDronesSee. The proposed ZoomDet is architecture-independent and can be applied to an arbitrary object detection architecture. Remarkably, on the SeaDronesSee dataset, ZoomDet offers more than 8.4 absolute gain of mAP with a Faster R-CNN model, with only about 3 ms additional latency. The code is available at https://github.com/twangnh/zoomdet_code.

cs.CV↗