arXiv subjects

Jingyuan Zhang

Publications and source records attributed to Jingyuan Zhang.

At least 19 recordsLinked to original sources

arXiv2026

PointEvent: Rethinking Event-based Tiny Object Detection via Serialized Motion Evidence Accumulation

Event cameras offer high temporal resolution and motion sensitivity for tiny UAV detection, yet distant targets generate sparse and fragmented events that are easily overwhelmed by clutter and ego-motion. Existing methods mainly rely on dense event representations or local sparse spatiotemporal modeling, resulting in redundant computation or fragmented modeling of motion continuity across distant asynchronous events. To address this limitation, we introduce serialized motion evidence accumulation, which treats motion continuity as an ordered evidence propagation process. Specifically, the same event stream is organized into locality-preserving spatiotemporal paths and chronology-preserving temporal paths through the latent complementary serializations. Based on this principle, we propose PointEvent, a lightweight event-wise state-space framework that alternates serialized scans across the complementary orders, progressively consolidating fragmented motion evidence beyond fixed local neighborhoods. A high-resolution event branch preserves fine-grained target responses, while compact context modulation suppresses interference. Experiments demonstrate that PointEvent achieves SOTA with the fewest parameters and fastest measured inference among the compared methods. Code: https://github.com/wzz-z/PointEvent

cs.CV↗

arXiv2026

HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?

As agents move from research prototypes to deployed tools, their capability increasingly depends on model-external execution infrastructure, commonly termed the agent harness. Changing this harness while holding model weights fixed can substantially alter task performance. Current agent evaluations typically report downstream performance under a chosen harness, leaving a model's ability to develop the harness itself comparatively underexplored. We introduce HarnessDev, a benchmark that shifts the unit of evaluation from task outputs to runnable infrastructure. HarnessDev covers two stages. In Creation, the agent starts from a minimal seed and a small number of cases, then builds a complete execution system. In Evolution, it starts from its own created harness and iteratively revises it using downstream execution feedback, with the goal of improving benchmark performance. We then evaluate each constructed harness on capability (task success on held-out benchmarks) and efficiency (execution-token cost). The reported Creation results cover six creator LLMs, four domains, and five downstream benchmarks totaling 2,207 unique downstream instances, with hidden evaluation tasks withheld from development. We find that generated harnesses remain substantially behind mature human-engineered references on code and on search and research, while matching or exceeding the selected references on writing and machine-learning experimentation, with large variation in execution cost. Evolution produces some performance gains, but they are unstable and transfer only partially to held-out tasks. Experiments with a fixed runtime model further show that the gains depend strongly on the model executing the harness, indicating limited transfer across models.

cs.SE↗

arXiv2026

S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement?

Large language models (LLMs) increasingly interact with external environments and accumulate substantial behavioral experience, yet existing agent benchmarks largely evaluate them as fixed policies. It therefore remains unclear whether an agent can actively test its behavior, judge the resulting experience, and use that experience to improve future decisions. We introduce \textbf{S\textsuperscript{3}Gym}, an interactive benchmark for evaluating LLM self-improvement through three coupled capabilities: \textbf{Self-Testing}, \textbf{Self-Judging}, and \textbf{Self-Improvement}. S$^3$Gym separates permissive exploration from strict held-out evaluation and instantiates this protocol in seven text-based games with executable environment verifiers. We evaluate three pathways for incorporating interaction experience: direct History ICL, score-conditioned Summary Memory, and parameter Training. Our experiments reveal that self-improvement is neither automatic nor uniform. Context-level experience improves performance for several model--game pairs, but the most effective pathway depends strongly on the task structure: summaries are beneficial when experience can be compressed into reusable strategic rules, yet often underperform raw history when success depends on precise, state-contingent information. Parameter training produces substantial gains on some tasks, but also exhibits unstable improvement and severe negative transfer on others. These findings show that recognizing successful actions is insufficient; agents must also transform feedback into executable and transferable policies. S$^3$Gym provides a unified framework for diagnosing this process and identifying the bottlenecks that prevent agents from translating interaction experience into reliable self-improvement.

cs.CL↗

arXiv2026

Aspire: Can Models Self-Evolve from Vague Goals?

Many important forms of human learning begin with a vague goal, such as "become a better physicist" or "improve at research." Learners must interpret the goal, identify capability gaps, decide how to learn, and determine whether they have actually improved. In contrast, existing work on LLM self-evolution typically begins with tasks and evaluation metrics specified by humans, reducing self-evolution to optimizing an explicit objective rather than deciding what and how to learn. We introduce ASPIRE, a benchmark for vague-goal-driven self-evolution. ASPIRE provides only a natural-language capability goal while downstream evaluation tasks remain hidden. The agent must operationalize the goal by choosing data and update methods, constructing training and validation signals, and deciding when to evaluate. ASPIRE supports both model-weight and agent-harness evolution in a unified interactive environment and evaluates the resulting systems on a hidden, expert-authored set of 520 items spanning six goals. Our experiments show that vague goals redirect search effort toward goal interpretation. Current agents routinely complete training and harness-editing loops, but weight-level gains remain sparse and unstable, and the strongest evolved harness remains below the engineered Qwen-Agent reference. Agents often train on mismatched data and trust narrow self-evaluations, so local gains fail to transfer to hidden evaluation and continued search and training can erase earlier improvements.

cs.CL↗

arXiv2026

A Frequency-Space Terahertz Transceiver Chip for Multi-Agent Communications and Spatial Awareness

Future indoor embodied-intelligence systems require scalable hardware platforms that support both high-capacity multi-agent connectivity and mutual spatial awareness. The terahertz (THz) spectrum offers abundant bandwidth and inherent spatial selectivity for integrated sensing and communication (ISAC); however, conventional phased arrays and programmable metasurfaces rely on dense beamforming networks, element-level control, or external THz illumination, making scalable multibeam operation challenging. Here, we report a fully integrated 208-258GHz 65-nm CMOS THz transceiver chip that monolithically integrates broadband front ends with heterogeneous leaky-wave metasurface (HLM) apertures within a 1.5mm by 4.9mm area. The HLM generates strongly dispersive leaky modes, enabling 75 degree frequency-controlled beam scanning with only four meta-atoms. Co-design of frequency-domain and spatial-domain mixing achieves spectrally clean frequency-to-space mapping for spatial-frequency division multiple access (SFDMA) communication. The THz chip demonstrates multi-agent simultaneous transmission and reception, two-dimensional localization, and sensing-enhanced communication, providing a scalable hardware platform for future THz embodied-intelligence networks.

physics.app-ph↗

arXiv2026

BashCoder-R1: Towards Robust and Explainable Bash Code Generation with Robustness-Aware Group Relative Policy Optimization

Bash scripts are critical for system administration, DevOps, and CI/CD, where code quality affects stability and security. However, LLM-generated scripts often lack reasoning and contain robustness flaws such as mishandled edge cases and unchecked failures. This is particularly critical in production environments where even minor errors can lead to service disruptions. We propose BashCoder-R1, a framework that jointly addresses both issues by treating explainability as a design goal. The pipeline has three stages. Continual Pre-training adapts to Bash syntax. Long Chain-of-Thought Supervised Fine-Tuning on expert-validated samples teaches risk-averse reasoning before code generation. Robustness-Aware Group Relative Policy Optimization optimizes a weighted reward for syntax correctness, robustness (verified by shellcheck), and format adherence. This staged design ensures that the model progressively acquires syntax knowledge, reasoning capability, and robust decision-making. On our BashBench benchmark (952 real-world tasks, 773 single-line and 179 multi-line), BashCoder-R1 achieves SyntaxPass of 100.00/94.97, RobustWarnRate of 4.01/16.47, RobustPass of 95.99/79.33, FuncRate of 93.01/93.85, and FullRate of 90.04/73.18 for single-line and multi-line tasks, respectively. These are relative FullRate improvements of 37.82 and 20.18 percent over the strongest baseline, DeepSeek-V3.2 (Reasoning). Human evaluation confirms its reasoning chains are highest in quality.

cs.SE↗

arXiv2026

Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection

Zero-shot visual anomaly detection has achieved remarkable progress, with recent vision-only approaches further improving performance while simplifying the inference pipeline. However, existing methods typically perform dense computation over all images and spatial tokens, despite the fact that normal samples dominate real-world scenarios and anomalies usually occupy only small regions. Token pruning offers a promising solution, but introduces an asymmetric pruning risk in anomaly detection: retaining normal tokens mainly incurs redundant computation, whereas removing anomalous tokens may eliminate the only evidence for detection and localization. This risk is particularly severe in early layers, where pruning provides the greatest computational benefit but anomaly semantics remain unreliable. We propose KeepAD, a defect-preserving token pruning framework that formulates token selection as high-recall, anomaly-aware routing. In shallow layers, KeepAD combines coverage-preserving selection over local $2\times2$ patch neighborhoods with deterministic anomaly rescue to reduce the risk of discarding subtle defects. In deeper layers, frozen normal and abnormal prototypes guide pruning under an image-adaptive token budget, aggressively removing low-risk normal tokens while preserving local anomaly evidence. Dense-to-sparse self-distillation further supervises early token routing without introducing additional inference overhead. Experiments on six industrial and seven medical zero-shot anomaly detection benchmarks show that KeepAD reduces the token retention ratio to below $20\%$, while limiting the average degradation in image-level and pixel-level AUROC to within $2.7$ percentage points. At the most aggressive operating point, KeepAD achieves a $7.9\times$ speedup over the strongest CLIP-based baseline.

cs.CV↗

arXiv2026

LP-SFT: Local-Preserving Supervised Fine-Tuning via Multimodal Entropy Structure

Supervised fine-tuning (SFT) is the standard approach for adapting pretrained language models to downstream domains, yet it often improves target-domain behavior at the cost of degrading pre-existing capabilities. Standard cross-entropy fine-tuning promotes only the observed label token and leaves unconstrained how probability mass is redistributed over other plausible alternatives, potentially distorting the rich local preference structure learned during pretraining. We first analyze next-token predictions using Shannon and Renyi entropies, revealing that pretrained models exhibit a regular multimodal entropy structure. These entropy peaks correspond to varying numbers of plausible alternatives, indicating that the base model intrinsically encodes rich distributional knowledge beyond the single supervised token. Motivated by this observation, we propose LP-SFT, a Local-Preserving Supervised Fine-Tuning objective designed to explicitly protect this inherent entropy structure. At each step, LP-SFT constructs a local top-K support of alternative tokens from the frozen base distribution. Crucially, it removes the supervised target token from this set to avoid conflicting with the cross-entropy objective, and applies a locally normalized KL divergence to maintain the base model's relative preference structure among the remaining non-label alternatives. Across mixed-domain and single-domain fine-tuning experiments, LP-SFT improves overall performance over vanilla SFT and recent SFT-enhancement baselines, achieving the best balance between pass@1 accuracy and pass@k performance. These results suggest that local preservation helps mitigate capability degradation without collapsing sampling-accessible diversity.

cs.CL↗

arXiv2026

ASUMOT: Motion-Consistency-Based Asynchronous UAV Detection and Tracking with Event Cameras

Event cameras offer microsecond-level temporal resolution and high dynamic range for low-altitude UAV perception. However, long-range UAVs often produce sparse, fragmented, and noise-contaminated event responses, where one semantic target may appear as multiple spatially separated blobs. Direct blob-level asynchronous tracking therefore suffers from duplicate trajectories and unstable identities. We propose ASUMOT, a motion-consistency-based asynchronous UAV detection and tracking framework operating directly on raw events. ASUMOT models each UAV as a set of motion-consistent event blobs. A local motion-consistency estimator triggers reliable candidates, a lightweight multi-task verifier provides UAV confidence and motion-direction cues, and motion-consistency clustering aggregates fragmented blobs into identity-consistent UAV tracks. We also introduce ES-UAV, a high-definition event-level UAV benchmark with dense semantic annotations. Experiments on public UAV tracking data and ES-UAV show that ASUMOT improves the accuracy--efficiency trade-off while preserving asynchronous event processing. Code and Dataset will be released.

cs.CV↗

arXiv2026

EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments

Pretraining scaling laws reveal that model capability improves predictably with data and compute. But learning from real world environments after deployment remains far less understood. Analyzing roughly 38,000 hours of agent interaction with the environment across 134 real world tasks, we find, to the best of our knowledge, the first evidence that overall performance during environment learning follows a log-sigmoid scaling law with remarkably high precision, reaching R^2 = 0.998. Across model generations, we also find that agent learning speed roughly doubles every three months. This discovery stems from EdgeBench, a suite of 134 real world tasks with ultra-long horizons, spanning scientific discovery, software engineering, combinatorial optimization, professional knowledge work, formal mathematics, and interactive games. Each task sustains at least 12 hours of continuous agent operation under rich, multilevel feedback, and is built through substantial expert effort. We publicly release 51 tasks and our full evaluation framework to accelerate the study of how agents learn from real world experience.

cs.CL↗

arXiv2026

Bash-Commenter: Leveraging Syntax-Aware Preference Optimization to Reinforce Large Language Model for Bash Code Comment Generation

Bash script comprehension is challenging due to Bash's syntactic freedom and complex command structures. Despite its critical role in system administration, Bash scripts often lack adequate comments, hindering readability and maintainability. Existing automated comment generation approaches face two main challenges: (1) limited training datasets that inadequately represent real-world Bash usage patterns; and (2) insufficient understanding of Bash-specific concepts by Large Language Models (LLMs). To address these, we propose Bash-Commenter, an advanced comment generation method based on LLaMA-3.1-8B. First, we construct a comprehensive dataset of complex, multi-line Bash scripts with high-quality comments. Second, we conduct Continual Pre-training (CPT) on large-scale Bash data, followed by Supervised Fine-tuning (SFT), strengthening the model's foundational knowledge of Bash syntax and semantics. Finally, we introduce Syntax-Aware Preference Optimization (SAPO), which constructs preference pairs by applying atomic operations to a script's Abstract Syntax Tree (AST), creating minimal pairs of correct and subtly incorrect scripts for fine-grained semantics learning. Our method outperforms state-of-the-art baselines, achieving 33.40% BLEU-4, 58.26% METEOR, and 57.03% ROUGE-L for 1,064 single-line commands, and 22.15% BLEU-4, 43.89% METEOR, and 32.80% ROUGE-L for 1,046 multi-line scripts. Human and LLM evaluations further confirm superior comment quality in correctness, completeness, and naturalness.

cs.SE↗

arXiv2026

Stochastic Optimal Control Sampling for Diffusion Inverse Problems

Benefiting from the strong ability to capture data distributions, diffusion models have become powerful tools for solving image inverse problems. The key is to controllably steer the sampling trajectory toward the measurements while respecting the diffusion prior. In this work, we introduce Stochastic Optimal Control Sampling (SOCS), which models the denoising process as a dynamical system and injects control signals via SOC. Previous SOC-based approach addresses inverse problems by optimizing over the entire trajectory, which is computationally expensive. In contrast, we derive a closed-form control update and apply it at each sampling step, pulling the measurement-consistent clean prediction back onto the denoising flow. In SOCS, we can readily modulate the control strength to align with the diffusion model's native capabilities and thereby enhance perceptual quality. Our method is compatible with a variety of linear stochastic differential equation backbones. Extensive experiments across a broad spectrum of image inverse tasks demonstrate that SOCS achieves accurate measurement-aligned reconstructions with improved visual fidelity and stronger quantitative performance.

cs.CV↗

arXiv2026

SAGE: Retain-Aware Post-Hoc Sanitization of Final Unlearning Vector

Large Language Model (LLM) unlearning aims to remove undesirable knowledge or behaviors while preserving retained capabilities. Current unlearning methods all involve a trade-off between unlearning and retention. We have found that the retention activation bias can also be used to quantify the damage an unlearning method inflicts on retention, without considering the specific implementation of the unlearning process. This allows us to restore retention performance for any unlearning method using a post-hoc approach. Therefore, we propose a complementary post-hoc setting to sanitize the final update vector without rerunning the original unlearning pipeline. In this setting, we design SAGE, Spectral Activation-GEometry Sanitization, a source-agnostic correction for final unlearning updates. SAGE collects real module inputs from a small retain proxy, extracts their dominant activation geometry, and solves a source-anchored optimization objective in closed form, which suppresses update components aligned with high-energy retained directions while preserving the source method's forgetting carrier. Across multiple unlearning methods, model scales, and benchmarks, SAGE consistently relieves the retain-forget trade-off, identifying post-hoc sanitization of final vectors as a practical and underexplored axis for machine unlearning.

cs.LG↗

arXiv2026

CVE-Factory: Scaling Expert-Level Agentic Tasks for Code Security Vulnerability

Evaluating and improving the security capabilities of code agents requires high-quality, executable vulnerability tasks. However, existing works rely on costly, unscalable manual reproduction and suffer from outdated data distributions. To address these, we present CVE-Factory, the first multi-agent framework to achieve expert-level quality in automatically transforming sparse CVE metadata into fully executable agentic tasks. Cross-validation against human expert reproductions shows that CVE-Factory achieves 95\% solution correctness and 96\% environment fidelity, confirming its expert-level quality. It is also evaluated on the latest realistic vulnerabilities and achieves a 66.2\% verified success. This automation enables two downstream contributions. First, we construct LiveCVEBench, a continuously updated benchmark of 190 tasks spanning 14 languages and 153 repositories that captures emerging threats including AI-tooling vulnerabilities. Second, we synthesize over 1,000 executable training environments, the first large-scale scaling of agentic tasks in code security. Fine-tuned Qwen3-32B improves from 5.3\% to 35.8\% on LiveCVEBench, surpassing Claude 4.5 Sonnet, with gains generalizing to Terminal Bench (12.5\% to 31.3\%). We open-source CVE-Factory, LiveCVEBench, Abacus-cve (fine-tuned model), training dataset, and leaderboard. All resources are available at https://github.com/livecvebench/CVE-Factory .

cs.CR↗

arXiv2026

SafeSteer: A Decoding-level Defense Mechanism for Multimodal Large Language Models

Multimodal large language models (MLLMs) are gaining increasing attention. Due to the heterogeneity of their input features, they face significant challenges in terms of jailbreak defenses. Current defense methods rely on costly fine-tuning or inefficient post-hoc interventions, limiting their ability to address novel attacks and involving performance trade-offs. To address the above issues, we explore the inherent safety capabilities within MLLMs and quantify their intrinsic ability to discern harmfulness at decoding stage. We observe that 1) MLLMs can distinguish the harmful and harmless inputs during decoding process, 2) Image-based attacks are more stealthy. Based on these insights, we introduce SafeSteer, a decoding-level defense mechanism for MLLMs. Specifically, it includes a Decoding-Probe, a lightweight probe for detecting and correcting harmful output during decoding, which iteratively steers the decoding process toward safety. Furthermore, a modal semantic alignment vector is integrated to transfer the strong textual safety alignment to the vision modality. Experiments on multiple MLLMs demonstrate that SafeSterr can improve MLLMs' safety by up to 33.40\% without fine-tuning. Notably, it can maintain the effectiveness of MLLMs, ensuring a balance between their helpfulness and harmlessness.

cs.AI↗

arXiv2026

DPN-LE: Dual Personality Neuron Localization and Editing for Large Language Models

With the widespread adoption of large language models (LLMs), understanding their personality representation mechanisms has become critical. As a novel paradigm in Personality Editing, most existing methods employ neuron-editing to locate and modify LLM neurons, requiring changes to numerous neurons and leading to significant performance degradation. This raises a fundamental question: Are all modified neurons directly related to personality representation? In this work, we investigate and quantify this specificity through assessments of general capability impact and representation-level patterns. We find that: 1) Current methods can change personalities but reduce overall performance. 2) Neurons are multifunctional, connecting personality traits and general knowledge. 3) Opposing personality traits demonstrate distinctly mutually exclusive representation patterns. Motivated by these findings, we propose DPN-LE (Dual Personality Neuron Localization and Editing), which identifies personality-specific neurons by contrasting MLP activations between high-trait and low-trait samples. DPN-LE constructs layer-wise steering vectors and applies dual-criterion filtering based on Cohen's $d$ effect size and activation magnitude to isolate mutually exclusive neuron subsets. Sparse linear intervention on these neurons enables precise personality control at inference time. Using only 1,000 contrastive sample pairs per trait, DPN-LE intervenes on $\sim$0.5\% of neurons while achieving competitive personality control and substantially better capability preservation across reasoning tasks. Experiments on LLaMA-3-8B-Instruct and Qwen2.5-7B-Instruct demonstrate the effectiveness and generalizability of our approach.

cs.CL↗

arXiv2026

WebTestBench: Evaluating Computer-Use Agents towards End-to-End Automated Web Testing

The emergence of Large Language Models (LLMs) has catalyzed a paradigm shift in programming, giving rise to "vibe coding", where users can build complete projects and even control computers using natural language instructions. This paradigm has driven automated webpage development, but it introduces a new requirement about how to automatically verify whether the web functionalities are reliably implemented. Existing works struggle to adapt, relying on static visual similarity or predefined checklists that constrain their utility in open-ended environments. Furthermore, they overlook a vital aspect of software quality, namely latent logical constraints. To address these gaps, we introduce WebTestBench, a benchmark for evaluating end-to-end automated web testing. WebTestBench encompasses comprehensive dimensions across diverse web application categories. We decompose the testing process into two cascaded sub-tasks, checklist generation and defect detection, and propose WebTester, a baseline framework for this task. Evaluating popular LLMs with WebTester reveals severe challenges, including insufficient test completeness, detection bottlenecks, and long-horizon interaction unreliability. These findings expose a substantial gap between current computer-use agent capabilities and industrial-grade deployment demands. We hope that WebTestBench provides valuable insights and guidance for advancing end-to-end automated web testing. Our dataset and code are available at https://github.com/friedrichor/WebTestBench.

cs.SE↗

arXiv2026

SQL-Commenter: Aligning Large Language Models for SQL Comment Generation with Direct Preference Optimization

SQL query comprehension is a significant challenge due to complex syntax, diverse join types, and deep nesting. Many queries lack adequate comments, severely hindering code readability, maintainability, and knowledge transfer. Automated SQL comment generation faces two main challenges: limited datasets that inadequately represent complex real-world queries, and Large Language Models' (LLMs) insufficient understanding of SQL-specific semantics. Our empirical analysis shows that even after continual pre-training and supervised fine-tuning, LLMs struggle with complex SQL semantics, yielding inaccurate comments. To address this, we propose SQL-Commenter, an advanced method based on LLaMA-3.1-8B. We first construct a comprehensive dataset of complex SQL queries with expert-verified comments. Next, we perform continual pre-training on a large SQL corpus to enhance the LLM's syntax and semantic understanding, followed by supervised fine-tuning. Finally, we introduce Direct Preference Optimization (DPO) using human feedback. SQL-Commenter utilizes a preference-based loss function to favor preferred outputs, enhancing fine-grained semantic learning and context-dependent quality assessment. Evaluated on the Spider and Bird benchmarks, SQL-Commenter significantly outperforms state-of-the-art baselines. On average, it surpasses the strongest baseline (Qwen3-14B) by 9.29, 4.99, and 13.23 percentage points on BLEU-4, METEOR, and ROUGE-L, respectively. Moreover, human evaluation demonstrates the superior quality of comments generated by SQL-Commenter in terms of correctness, completeness, and naturalness.

cs.SE↗