Searcharxiv⌕ Search

arXiv subjects

Jiaming Zhang

Publications and source records attributed to Jiaming Zhang.

At least 19 recordsLinked to original sources
arXiv2026

Efficient Deep Learning Adaptation for Cross-Environment RSS-Based Indoor Localization

Received signal strength (RSS)-based indoor localization has attracted increasing attention due to its low cost and compatibility with existing wireless infrastructures. However, RSS measurements are highly sensitive to environmental variations, making it challenging for deep learning-based localization models to generalize across different physical configurations. Driven by these challenges, this paper proposes a scenario-adaptive RSS localization framework based on backbone reuse and efficient adaptation. The proposed architecture consists of a lightweight extractor and a shared backbone, where the extractor projects heterogeneous RSS inputs into a unified feature space and the backbone captures transferable localization knowledge. During adaptation, the backbone trained from the source dataset is reused, while the extractor is optimized to adapt to the new dataset configuration. A short optimization stage is further introduced to slightly refine the backbone with a lower learning rate. Experimental results on four datasets demonstrate that the proposed training strategy enables faster convergence and improved adaptation compared with training from scratch. In addition, the proposed framework reduces training time under different dataset configurations, verifying its effectiveness and efficiency for adaptive RSS-based indoor localization.

eess.SP↗
arXiv2026

In-Place Instruction Following in Diffusion Language Models

Diffusion Large Language Models (dLLMs) generate text via bidirectional iterative denoising, naturally supporting user-specified constraints anchored at arbitrary output positions, a paradigm known as In-place Prompting (IPP). We formalize this as the In-place Instruction Following (IIF) task and construct IIF-Bench, a hierarchical benchmark spanning literal, style, and discourse-function constraints, paired with a rubric-based local-global evaluation protocol. An inference-time attention-bias probe suggests that vanilla dLLMs often under-prioritize constraint spans during denoising. We then propose GRAFT, an IPP-oriented post-training framework combining constraint-aware SFT and preference optimization. On four representative dLLMs, GRAFT raises the average IIF score from 57.75 to 73.10 (+15.35 points), with absolute gains of 15.91 and 15.57 points on literal and discourse-function constraints, while preserving general generation ability.

cs.CL↗
arXiv2026

Benchmarking Vision-Language Models for Automated Pathology Diagnosis and Report Generation

The rapid advancement of vision-language models (VLMs) has accelerated progress in computational pathology; however, whole-slide image (WSI)-based pathology report generation remains limited by the scarcity of large-scale WSI--report datasets and the complexity of mapping spatially distributed visual patterns to structured clinical text. To address this, we introduce a clinically curated Pan-Asia WSI--report dataset of approximately 10,500 pairs from five institutions and establish the REG 2025 benchmark through a MICCAI challenge for systematic evaluation of multimodal models. We analyze submitted methods spanning pretrained VLMs, multiple-instance learning frameworks, hierarchical expert models, retrieval-augmented generation, and cross-modal Transformers. Rather than indicating that VLM use alone was sufficient for superior performance, the results suggest that top-performing methods benefited from structured report representations, hierarchical diagnostic decomposition, and effective multimodal grounding. We identify key limitations, including instability in quantitative attribute estimation (e.g., numeric hallucination) and a tendency toward diagnostic overspecification, with some errors resembling known diagnostic pitfalls in routine pathology. These findings establish REG 2025 as a benchmark for evaluating WSI-based structured report generation and vision-language understanding in computational pathology, providing insights for the design of clinically grounded multimodal pathology models.

cs.CV↗
arXiv2026

X$^2$Localizer: Cross-grained Alignment for Progressive Cross-view Video Geo-localization

Cross-view Video Geo-localization (CVG) aims to localize ground-view videos by retrieving their corresponding geo-tagged aerial images. However, CVG approaches rely on fixed-length inputs and post-hoc refinement, hindering online-oriented localization under partial or dynamic observations. In this work, we formulate Progressive Cross-view Video Geo-localization (PCVG) as a deployment-oriented extension and evaluation protocol of CVG, enabling localization under varying temporal budgets, prefix-based inference, random-start evaluation, and long-range localization with interruptions. To explore PCVG, we introduce X$^2$Localizer, a cross-grained alignment framework that jointly supervises global prefix-to-aerial retrieval and token-aggregated frame--aerial-tile matching with a budget-dependent asymmetric objective. Furthermore, we introduce a Sliding-Window Re-Localization (SWRL) strategy that dynamically refreshes candidate regions for failure recovery and long-range deployment without full-sequence reprocessing. Extensive experiments show that X$^2$Localizer preserves conventional full-video performance, with marginal gains of +0.1 Recall@1 and +0.3 Recall@10, while substantially improving early localization. In the challenging single-frame setting, X$^2$Localizer improves coarse retrieval by +4.7 Recall@1 and +11.5 Recall@10 over the previous state-of-the-art method. With SWRL, our approach further enables robust progressive localization under random-start and long-distance scenarios, narrowing the gap between benchmark evaluation and real-world deployment.

cs.CV↗
arXiv2026

SERS study of single-live-cell electrical permeabilization dynamics via plasmonic nanotubes

There is a growing demand for minimally invasive methods to analyze intracellular processes and signaling activities in individual living cells, including the identification of tumorigenic cell subpopulations. However, most conventional analytical methods require cell lysis, precluding repeated measurements in the same cell over time, or rely on exogenous labels and reporters that may perturb cellular function. Various applications based on vertical nanotubes have been developed that enable live cell monitoring and analysis by electroporation with low voltages. However, the extent and duration of membrane permeability and kinetics of membrane repair remain elusive. Here, we built a plasmonic platform with the capacity of surface enhanced Raman spectroscopy (SERS) to monitor the electroporation-induced membrane permeability dynamics in individual live cells attached onto 100-nm diameter nanotubes of 2 um height. Fibronectin was employed as extracellular matrix (ECM)-coating to facilitate cell attachment onto nanotubes. Using fluorescent-dye delivery as an independent validation method, we show that the fabricated nanostructures induce localized electrical permeabilization of the plasma membrane and enable monitoring of its subsequent recovery. We further use SERS to track molecular changes at the membrane during permeabilization and resealing. The SERS spectra provide molecular-level insight into changes in membrane-associated components and the ECM during electroporation and subsequent membrane recovery. Real time monitoring of pulse induced molecular changes holds great promise for characterizing intracellular signaling, cellular states, and cellular heterogeneity at the single cell level, including the identification of tumorigenic subpopulations. This capacity could facilitate the development of novel biosensing assay.

physics.app-ph↗
arXiv2026

SeqLoc: Beyond the Single Frame for Cross-View Geo-Localization in Feature-Sparse Scenes

Cross-View Geo-Localization (CVGL) with OpenStreetMap (OSM) performs well in structure-rich urban environments but collapses in feature-sparse scenes such as rural roads. To study this failure mode, in this work, we introduce CV-FSS, a benchmark that pairs sequential panoramas from five rural regions with aligned OSM maps, on which single-frame methods degrade drastically. We then propose SeqLoc, an online test-time sequence aggregation mechanism that recursively maintains a log-belief volume with three key components: (1) Entropy-Tempered Uncertainty (ETU) tempers each incoming pose likelihood volume by its normalized entropy; (2) Map-Guided Relocalization (MGR) mixes a map-shaped recovery distribution into the belief so that a suppressed true pose can recover; (3) Peak-Anchored Smoothing (PAS) derives the final pose at sub-grid precision. Extensive experiments on CV-FSS and CV-RHO demonstrate that SeqLoc outperforms single-frame localization by a large margin, improving both position and orientation recall by over 50%. The benchmark and source code are publicly available at https://zhengjunwei.com/publications/SeqLoc/SeqLoc.html.

cs.CV↗
arXiv2026

SynChain: Inducing Computer-Use Agent Systems to Construct Their Own Attack Chains

Computer-use agents~(CUAs) have transformed large language models into persistent execution systems capable of generating, storing, and reusing artifacts like skills and memory entries. However, existing security defenses largely treat attacks as externally triggered or temporally bounded, leaving a critical gap in addressing how compromise can propagate internally through an agent's own persistent state. We reveal that malicious influence can be covertly embedded into the structural redundancies of autonomously synthesized artifacts, allowing it to survive internal state updates and bypass standard vetting mechanisms. To formalize this threat, we introduce SynChain, a self-synthesized attack paradigm utilizing persistence-aware directed supervised fine-tuning to induce agents to create poisoned yet benign-looking artifacts. To systematically evaluate this propagation, we construct CUAChain, a dataset comprising 30 benign task chains and three attack objectives. SynChain enables dormant payloads to seamlessly reactivate in future workflows as trusted context, operating entirely without new malicious exogenous inputs. Extensive experiments on OpenClaw, Codex, and Claude Code under four defense settings demonstrate that SynChain achieves high attack success and outperforms adapted baselines, proving that securing CUAs requires provenance-aware reasoning over cross-task execution trajectories.

cs.CR↗
arXiv2026

Adversarial Attacks for Good: A Survey of Proactive Protection across the Visual Content Lifecycle

Once visual content enters an AI pipeline, its owner often retains little technical control over how it is used. Legal and regulatory remedies can address misuse, but many technical interventions must be applied earlier, when content is released or accessed. This survey examines the protective paradigm that has grown around this intervention point, which we call \emph{adversarial attacks for good}. Perturbations and structured signals long studied as attacks on learned models are instead applied by data owners, creators, platforms, or auditors to disrupt unauthorized automation or support later accountability. Five research communities have arrived at this inversion largely independently, each addressing a different stage of a visual asset's lifecycle: privacy filters against unwanted recognition at sharing time, unlearnable examples against unauthorized training, generative safeguards against malicious editing or imitation, adversarial CAPTCHAs for access control against automated agents, and provenance mechanisms for post-circulation attribution. Although developed in separate venues with incompatible success criteria, many of these methods exploit persistent gaps between human perception, semantic interpretation, and machine inference, suggesting that the paradigm remains relevant as visual pipelines evolve toward multimodal models and autonomous agents. To make their claims comparable, we evaluate all five families along shared axes of transferability, adaptability, and deployment readiness. Across the lifecycle, we find that most protections are still validated mainly against static or weakly adaptive adversaries, while evidence beyond controlled benchmarks remains scarce. We close by consolidating cross-stage countermeasures and open problems for robust, composable, and deployable owner-side protection.

cs.CR↗
arXiv2026

Generative Adversarial Reconstruction with Adaptive Thresholding for Obstructed Targets in Computational Microwave Imaging

In this work, an end-to-end generative adversarial framework for computational microwave imaging (CMI) is proposed to reconstruct the targets of interest directly from the measurements of targets obstructed by undesired objects. It integrates a conditional generative adversarial network (cGAN) with a learnable soft-threshold module (STM) to adaptively suppress non-target related information. The proposed framework is evaluated on a diverse dataset comprising MNIST digits obstructed by E-MNIST letters for training and testing, as well as on measurements acquired with an experimental CMI system. In addition, further studies involving objects with different geometries are conducted, demonstrating that the proposed approach can be adapted to other types of objects. Numerical experiments show that the proposed cGAN-STM achieves a normalized mean square error (NMSE) of 0.066 and a structural similarity index (SSIM) of 0.876 under ideal conditions. Comprehensive analyses, including benchmarking, analysis of the STM mechanism, and evaluation under different obstruction sizes, are also conducted. The performance of the model under different signal-to-noise ratio (SNR) scenarios is also evaluated, achieving reasonable reconstruction quality at 15 dB SNR with an NMSE of 0.168 and an SSIM of 0.700. Even at low SNR levels, recognizable target outlines are preserved. These results highlight the effectiveness and adaptability of the proposed method.

eess.SP↗
arXiv2026

PercepCap: Video Captioner with Structured Spatio-Temporal Perception

Video captioning requires fine-grained spatio-temporal understanding of videos, including spatial perception of where objects are located and temporal perception of when events occur. Existing MLLMs usually generate captions directly from video inputs without exposing the perceptual evidence behind descriptions. As a result, mistakes in spatiotemporal perception are only observed in the final caption, making it difficult to identify the underlying perceptual errors directly. To address these issues, we present PercepCap, a perception-aware video captioning framework that makes perceptual evidence explicit before producing the final caption. Specifically, PercepCap follows a perceive-describe generation chain, where the model first produces a spatiotemporal perception trace comprising object trajectories and temporal events, and then generates the final caption conditioned on the perceived evidence. To support this, we design a two-stage training strategy. Perceive-then-Describe Supervised Fine-tuning adapts the model from caption-only generation to the proposed perceive-describe chain, while Perception-Grounded Reinforcement Learning optimizes perception trace and caption quality with joint rewards over perception chain and the final caption. To support our two-stage training, we introduce Caption-Anchored Perception Data Construction. This pipeline builds the SFT and RL training data by first generating a caption-only description, extracting the objects and events it mentions, and grounding them back in the video with boxes and timestamps. This yields caption-aligned perception data that provides solid training ground truth, ensuring that the explicit perception trace and final caption refer to the same objects and events. Across direct caption and caption-to-QA evaluation, PercepCap consistently improves upon the Qwen3-VL baseline and demonstrates leading caption quality.

cs.CV↗
arXiv2026

LazyMem: Retrieve Broadly, Construct Selectively for Efficient Long-Term Agent Memory

Long-term memory enables LLM agents to leverage past interactions, but dialogue histories quickly exceed the context window, forcing agents to retrieve relevant subsets at query time. Because useful evidence is sparse and scattered across verbose conversations, retrieval faces a fundamental tension: broadening recall improves coverage but floods downstream reasoning with noise, while compressing memories at write time eases retrieval but irreversibly discards details that future queries may need. We introduce LazyMem, which resolves this tension by deferring all memory construction to query time. Given a retrieved candidate pool, a lightweight model processes it in overlapping parallel windows, selectively retaining and compressing only query-relevant content. The model is trained with supervised fine-tuning followed by reinforcement learning, using a reward that jointly encourages the identification of relevant messages and the generation of compressions that are faithful to the source and useful for answering the query. On LongMemEval, LazyMem-4B achieves an LLM-judge accuracy of 0.85, outperforming the strongest non-oracle baseline while using only 213 answer-context memory tokens, 21.0 times fewer than the baseline. It further generalizes to LoCoMo without target-domain training and reduces mean latency relative to the prior query-time baseline. Code is available at https://github.com/allacnobug/LazyMem.

cs.AI↗
arXiv2026

Point Tracking in Surgery--The 2025 Surgical Tattoos in Infrared Challenge (STIRC2025)

Point tracking in surgery is crucial to enable applications in downstream tasks such as segmentation, 3D reconstruction, virtual tissue landmarking, autonomous probe-based scanning, and subtask autonomy. This paper introduces the 2025 iteration of a point tracking challenge to address this, wherein participants submit their algorithms for quantification. Their algorithms are evaluated using a dataset named surgical tattoos in infrared (STIR), with the challenge named the STIR Challenge 2025 (STIRC2025). The STIR Challenge 2025 comprises two quantitative components: accuracy and efficiency. The accuracy component tests the accuracy of algorithms on in vivo and ex vivo sequences. The efficiency component tests algorithm inference latency. The challenge was conducted as a part of MICCAI EndoVis 2025, and seven teams participated in this challenge. In this paper we summarize the challenge results and participant methods. The challenge dataset is available at: https://zenodo.org/records/20191078, and the code for baseline models and metrics calculation is available here: https://github.com/athaddius/STIRMetrics

cs.CV↗
arXiv2026

Tri-Info: Generalizable, Interpretable Failure Prediction for VLA Models via Information Theory

Vision-Language-Action (VLA) models are increasingly deployed across diverse tasks, yet they remain black boxes whose physical interactions can cause irreversible harm, making generalizable and interpretable failure detection essential. We observe that successful and failed rollouts carry systematically different information-theoretic signatures. Building on this, we formalize VLA control as a closed-loop information pipeline and derive the Triple Information-theoretic (Tri-Info) signals that capture whether actions remain diverse, temporally consistent, and coupled to state transitions. Across six VLA models and three benchmark environments, Tri-Info matches the strongest baselines in-domain. Moreover, Tri-Info transfers across architectures, environments, and the sim-to-real gap without retraining, reaching 83\% accuracy on real-world tasks where prior detectors collapse to chance. This establishes Tri-Info as a simple yet powerful method that not only detects failures with strong cross-domain generalization, but also delivers interpretable diagnostics of the underlying failure modes.

cs.RO↗
arXiv2026

High-Degree-of-Freedom Lightweight Bioinspired Leg for Enhanced Mobility in Small Robots

In microrobotics, enhancing locomotion capabilities by increasing the degrees of freedom (DoF) of leg mechanisms under severe spatial constraints remains a significant challenge. Inspired by insect locomotion, this paper presents a novel micro-scale parallel leg mechanism with four degrees of freedom, and systematically analyzes its mechanical design, electrical system, and kinematics. The design incorporates two spherical five-bar linkages to achieve spatial motion within a parallel four-bar configuration. Furthermore, a concentric design strategy is employed to simplify the analytical solution of the leg kinematics. Due to the parallel system architecture, all actuators are located on the main body, substantially reducing the equivalent inertia of moving parts compared to traditional high-DOF leg structures. The total mass of the system is only 18.9 g, with an end-effector output force of approximately 0.5 N and a workspace exceeding 22255 mm3. Experimental results demonstrate that the proposed single-leg mechanism achieves excellent motion flexibility, highlighting its potential for micro bio-inspired robotics.

cs.RO↗
arXiv2026

Improved bounds for lines and $1$-separated sets in Euclidean Ramsey theory

Let $K$ be a $1$-separated set of diameter at most $R-1$, and let $\ell_m$ denote a collection of $m$ points on a line, with consecutive points of distance $1$ apart. Conlon and Fox (2019) demonstrated a coloring of $n$-dimensional Euclidean space avoiding red congruent copies of $\ell_2$ and blue congruent copies of $K$ for $|K| > 10000^n\log_2 R$. We show here a stronger bound, that in fact $|K| > (6.79 + o(1))^n\log R$ suffices for arbitrary $1$-separated $K$, while the improvement $|K| > (5 + o(1))^n\log R$ holds in many cases, including when $K = \ell_m$, or more generally when $K$ is contained in a low-dimensional affine subspace. We also make a special study of the case when $n=2$, demonstrating a two-coloring of two-dimensional Euclidean space avoiding red copies of $\ell_2$ and blue copies of $\ell_{6330}$. This latter result addresses a question of Erd\H{o}s and Graham.

math.CO↗
arXiv2026

How Proposal Novelty, Topical Diversity, and Theory-Practice Balance Shape Scholarly Outcomes in Funded Education Research

Education research occupies a distinctive position in public science because it is expected to advance scholarly knowledge while also informing learning, teaching, participation, and workforce development. This study examines how the intellectual characteristics of NSF-funded education proposals are associated with the subsequent academic performance of funded scholars. Linking 8,715 NSF education awards from 1990 to 2020 with 84,519 publications by principal investigators, the analysis focuses on four major NSF education divisions that collectively span undergraduate and graduate levels, formal and informal learning environments, and inclusive educational initiatives. Proposal novelty is measured as semantic distance from prior funded projects within the same division, topical diversity as breadth across latent research themes, and intellectual orientation as theoretical, practical, or balanced. The results show that NSF education funding is consistently associated with higher publication output across divisions. However, this increase is not accompanied by stronger citation performance or higher journal-level visibility; citation and CiteScore estimates are often negative, particularly in later decades. Proposal novelty shows limited and uneven associations with post-award outcomes, whereas topical diversity is more clearly related to publication growth in some divisions but weaker citation-based performance in others. Balanced proposals that integrate theoretical and practical aims display the most favourable overall profile, combining positive publication associations with fewer negative citation-based patterns. These findings highlight the importance of evaluating education research funding through multiple academic outcomes and division-specific research contexts.

cs.DL↗
arXiv2026

Demystifying the Optimal Fair Classifier in Multi-Class Classification

Ensuring fair and equitable treatment across diverse groups, particularly in multi-class classification tasks, poses a significant challenge due to the persistent biases inherent in machine learning models. Most existing bias mitigation techniques are tailored to binary settings, and the presence of multi-dimensional outputs and complex fairness mechanisms makes their extension to multi-class scenarios neither straightforward nor effective. In this paper, we investigate two fundamental, unresolved challenges in fair classification: (i) characterizing the optimal accuracy-fairness frontier in multi-class settings, and (ii) designing practical algorithms that attain this optimum in different training phases. To tackle these challenges, we first specify an analytically tractable probabilistic formulation of the optimal classifier under fairness constraints. Building upon this, we propose two attribute-blind algorithms to enforce fairness requirements in practice: an in-processing approach for fairness intervention during training via the reduction approach, and a post-processing approach for fine-tuning output probabilities with plug-in estimation. Theoretical analysis reveals that both methods converge to the optimal accuracy-fairness Pareto frontier. Experiments conducted on multiple datasets demonstrate the superior performance of our methods in balancing accuracy and fairness.

cs.LG↗
arXiv2026

Algorithmic Recourse of In-Context Learning for Tabular Data

As predictive models are increasingly deployed in high-stakes settings such as credit approval, there is a growing need for post-hoc methods that provide recourse to affected individuals. Many such models operate on tabular data, where features correspond to real-world attributes. Recently, in-context learning (ICL) has enabled large language models to perform tabular prediction by conditioning on labeled examples at inference time, without explicit training. However, algorithmic recourse for tabular decision-making under ICL remains largely unexplored. In this work, we present the first study of algorithmic recourse for tabular data under ICL. We carry out a theoretical analysis, showing that recourse remains well-defined and bounded, and we characterize how recourse converges toward classical solutions as the context size increases. In practice, we propose a novel zeroth-order recourse framework, Adaptive Subspace Recourse for In-Context Learning (ASR-ICL), that efficiently generates actionable and sparse recourse for black-box ICL models. The proposed framework naturally extends to multi-class tabular tasks. Experiments across multiple real-world datasets and models demonstrate that ASR-ICL achieves recourse quality comparable to existing methods with fewer queries and empirically confirm the predicted convergence behavior, supporting our theoretical analysis.

cs.LG↗