arXiv subjects

Nan Zhang

Publications and source records attributed to Nan Zhang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Vista: Scene-Aware Optimization for Streaming Video Question Answering under Post-Hoc Queries

Streaming video question answering (Streaming Video QA) poses distinct challenges for multimodal large language models (MLLMs), as video frames arrive sequentially and user queries can be issued at arbitrary time points. Existing solutions relying on fixed-size memory or naive compression often suffer from context loss or memory overflow, limiting their effectiveness in long-form, real-time scenarios. We present Vista, a novel framework for scene-aware streaming video QA that enables efficient and scalable reasoning over continuous video streams. The innovation of Vista can be summarized in three aspects: (1) scene-aware segmentation, where Vista dynamically clusters incoming frames into temporally and visually coherent scene units; (2) scene-aware compression, where each scene is compressed into a compact token representation and stored in GPU memory for efficient index-based retrieval, while full-resolution frames are offloaded to CPU memory; and (3) scene-aware recall, where relevant scenes are selectively recalled and reintegrated into the model input upon receiving a query, enabling both efficiency and completeness. Vista is model-agnostic and integrates seamlessly with a variety of vision-language backbones, enabling long-context reasoning without compromising latency or memory efficiency. Extensive experiments on StreamingBench demonstrate that Vista achieves state-of-the-art performance, establishing a strong baseline for real-world streaming video understanding.

cs.CV↗

arXiv2026

QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals

Weight-only quantization is important for compressing Large Language Models (LLMs). Inspired by the spirit of classical magnitude pruning, we study whether the magnitude of weight updates during reasoning-incentivized fine-tuning can provide valuable signals for quantizing Large Reasoning Models (LRMs). We hypothesize that the smallest and largest weight updates during fine-tuning are more important than those of intermediate magnitude, a phenomenon we term "protecting both ends". Upon hypothesis validation, we introduce QuantLRM, which stands for weight quantization of LRMs via fine-tuning signals. We fit simple restricted quadratic functions on weight updates to protect both ends. By multiplying the average quadratic values with the count of zero weight updates of channels, we compute channel importance that is more effective than using activation or second-order information. We run QuantLRM to quantize various fine-tuned models (including supervised, direct preference optimization, and reinforcement learning fine-tuning) over four reasoning benchmarks (AIME-120, FOLIO, temporal sequences, and GPQA-Diamond) and empirically find that QuantLRM delivers a consistent improvement for LRMs quantization, with an average improvement of 6.55% on a reinforcement learning fine-tuned model. Also supporting non-fine-tuned LRMs, QuantLRM gathers effective signals via pseudo-fine-tuning, which greatly enhances its applicability.

cs.LG↗

arXiv2026

Triage: Hierarchical Visual Budgeting for Efficient Video Reasoning in Vision-Language Models

Vision-Language Models (VLMs) face significant computational challenges in video processing due to massive data redundancy, which creates prohibitively long token sequences. To address this, we introduce Triage, a training-free, plug-and-play framework that reframes video reasoning as a resource allocation problem via hierarchical visual budgeting. Its first stage, Frame-Level Budgeting, identifies keyframes by evaluating their visual dynamics and relevance, generating a strategic prior based on their importance scores. Guided by this prior, the second stage, Token-Level Budgeting, allocates tokens in two phases: it first secures high-relevance Core Tokens, followed by diverse Context Tokens selected with an efficient batched Maximal Marginal Relevance (MMR) algorithm. Extensive experiments demonstrate that Triage improves inference speed and reduces memory footprint, while maintaining or surpassing the performance of baselines and other methods on various video reasoning benchmarks.

cs.CV↗

arXiv2026

A census of star-formation and gas mass tracers in two lensed $z \sim 4$ dusty star-forming galaxies

We present new and archival Atacama Large Millimeter/submillimeter Array (ALMA) observations of two strongly lensed dusty star-forming galaxies (DSFGs) selected from the South Pole Telescope survey, SPT0418-47 $(z = 4.225)$ and SPT2147-50 $(z = 3.760)$. We study the [C II], CO(7-6), [C I](2-1), and, in SPT0418-47, $p$-H$_2$O emission, which along with the underlying continuum (rest-frame 160 $μ$m and 380 $μ$m) are routinely used as tracers of gas mass and/or star-formation rate (SFR). We perform a pixel-by-pixel analysis of both sources in the image plane to study the resolved Kennicutt-Schmidt relation, finding generally good agreement between the slopes of the SFR versus gas mass surface density using the different tracers. Using lens modeling methods, we find that the dust emission is more compact than the line emission in both sources, with CO(7-6) and [C I](2-1) similar in extent and [C II] the most extended, reminiscent of recent findings of extended [C II] spatial distributions in galaxies at similar cosmic epochs. We develop the [C I](2-1) / CO(7-6) flux density ratio as an observable proxy for gas depletion timescale ($τ_{\rm dep}$), which can be applied to large samples of DSFGs, in lieu of more detailed inferences of this timescale which require analysis of observations at multiple wavelengths. Furthermore, the extended [C II] emission in both sources, compared to the total continuum and line emission, suggests that [C II], used in recent years as a molecular gas mass and SFR tracer in high-$z$ galaxies, may not always be a suitable tracer of these physical quantities.

astro-ph.GA↗

arXiv2025

A transfer learning approach for automatic conflicts detection in software requirement sentence pairs based on dual encoders

Software Requirement Document (RD) typically contain tens of thousands of individual requirements, and ensuring consistency among these requirements is critical for the success of software engineering projects. Automated detection methods can significantly enhance efficiency and reduce costs; however, existing approaches still face several challenges, including low detection accuracy on imbalanced data, limited semantic extraction due to the use of a single encoder, and suboptimal performance in cross-domain transfer learning. To address these issues, this paper proposes a Transferable Software Requirement Conflict Detection Framework based on SBERT and SimCSE, termed TSRCDF-SS. First, the framework employs two independent encoders, Sentence-BERT (SBERT) and Simple Contrastive Sentence Embedding (SimCSE), to generate sentence embeddings for requirement pairs, followed by a six-element concatenation strategy. Furthermore, the classifier is enhanced by a two-layer fully connected feedforward neural network (FFNN) with a hybrid loss optimization strategy that integrates a variant of Focal Loss, domain-specific constraints, and a confidence-based penalty term. Finally, the framework synergistically integrates sequential and cross-domain transfer learning. Experimental results demonstrate that the proposed framework achieves a 10.4% improvement in both macro-F1 and weighted-F1 scores in in-domain settings, and an 11.4% increase in macro-F1 in cross-domain scenarios.

cs.SE↗

arXiv2025

Attention-Guided Patch-Wise Sparse Adversarial Attacks on Vision-Language-Action Models

In recent years, Vision-Language-Action (VLA) models in embodied intelligence have developed rapidly. However, existing adversarial attack methods require costly end-to-end training and often generate noticeable perturbation patches. To address these limitations, we propose ADVLA, a framework that directly applies adversarial perturbations on features projected from the visual encoder into the textual feature space. ADVLA efficiently disrupts downstream action predictions under low-amplitude constraints, and attention guidance allows the perturbations to be both focused and sparse. We introduce three strategies that enhance sensitivity, enforce sparsity, and concentrate perturbations. Experiments demonstrate that under an $L_{\infty}=4/255$ constraint, ADVLA combined with Top-K masking modifies less than 10% of the patches while achieving an attack success rate of nearly 100%. The perturbations are concentrated on critical regions, remain almost imperceptible in the overall image, and a single-step iteration takes only about 0.06 seconds, significantly outperforming conventional patch-based attacks. In summary, ADVLA effectively weakens downstream action predictions of VLA models under low-amplitude and locally sparse conditions, avoiding the high training costs and conspicuous perturbations of traditional patch attacks, and demonstrates unique effectiveness and practical value for attacking VLA feature spaces.

cs.CV↗

arXiv2025

Efficient Community Detection Over Streaming Bipartite Networks (Technical Report)

The streaming bipartite graph is widely used to model the dynamic relationship between two types of entities in various real-world applications, including movie recommendations, location-based services, and online shopping. Since it contains abundant information, discovering the dense subgraph with high structural cohesiveness (i.e., community detection) in the bipartite streaming graph is becoming a valuable problem. Inspired by this, in this paper, we study the structure of the community on the butterfly motif in the bipartite graph. We propose a novel problem, named Community Detection over Streaming Bipartite Network (CD-SBN), which aims to retrieve qualified communities with user-specific query keywords and high structural cohesiveness at snapshot and continuous scenarios. In particular, we formulate the user relationship score in the weighted bipartite network via the butterfly pattern and define a novel $(k,r,σ)$-bitruss as the community structure. To efficiently tackle the CD-SBN problem, we design effective pruning strategies to rule out false alarms of $(k,r,σ)$-bitruss and propose a hierarchical synopsis to facilitate the CD-SBN processing. We develop efficient algorithms to answer snapshot and continuous CD-SBN queries by traversing the synopsis and applying pruning strategies. With extensive experiments, we demonstrate the performance of our CD-SBN approach on real/synthetic streaming bipartite networks.

cs.SI↗

arXiv2025

Asymptotically circularly polarized bound states in the continuum

We study a class of bound states in the continuum (BICs) in all-dielectric periodic structures, near which resonant states approach ideal circularly polarized states (CPSs). We term these BICs {\em asymptotically circularly polarized BICs} ({\em acp}-BICs) and identify two types: single-angle and all-angle. Single-angle {\em acp}-BICs permit convergence to left- or right-handed CPSs only along a single momentum-space direction, whereas all-angle {\em acp}-BICs exhibit convergence to CPSs of a single handedness throughout the entire momentum space, rendering them exceptionally promising for chiral optical applications. We reveal that the existence of {\em acp}-BICs is underpinned by total reflection of circularly polarized waves. Moreover, all-angle {\em acp}-BICs qualify as super-BICs, with uniform nearby polarization being an intrinsic property. In addition, a bifurcation theory is developed to analyze the emergence of genuine CPSs from {\em acp}-BICs under $C_{2}$-symmetric structural perturbations. Our results suggest {\em acp}-BICs as a platform for singular and chiral optical responses in all-dielectric systems.

physics.optics↗

arXiv2025

Kinematically Controllable Cable Robots with Reconfigurable End-effectors

To enlarge the translational workspace of cable-driven robots, one common approach is to increase the number of cables. However, this introduces two challenges: (1) cable interference significantly reduces the rotational workspace, and (2) the solution of tensions in cables becomes non-unique, resulting in difficulties for kinematic control of the robot. In this work, we design structurally simple reconfigurable end-effectors for cable robots. By incorporating a spring, a helical-grooved shaft, and a matching nut, relative linear motions between end-effector components are converted into relative rotations, thereby expanding the rotational workspace of the mechanism. Meanwhile, a bearing is introduced to provide an additional rotational degree of freedom, making the mechanism non-redundant. As a result, the robot's motion can be controlled purely through kinematics without additional tension sensing and control.

cs.RO↗

arXiv2025

Dynamic Service Scheduling and Resource Management in Energy-Harvesting Multi-access Edge Computing

Multi-access Edge Computing (MEC) delivers low-latency services by hosting applications near end-users. To promote sustainability, these systems are increasingly integrated with renewable Energy Harvesting (EH) technologies, enabling operation where grid electricity is unavailable. However, balancing the intermittent nature of harvested energy with dynamic user demand presents a significant resource allocation challenge. This work proposes an online strategy for an MEC system powered exclusively by EH to address this trade-off. Our strategy dynamically schedules computational tasks with dependencies and governs energy consumption through real-time decisions on server frequency scaling and service module migration. Experiments using real-world datasets demonstrate our algorithm's effectiveness in efficiently utilizing harvested energy while maintaining low service latency.

cs.DC↗

arXiv2025

PEEL: A Poisoning-Exposing Encoding Theoretical Framework for Local Differential Privacy

Local Differential Privacy (LDP) is a widely adopted privacy-protection model in the Internet of Things (IoT) due to its lightweight, decentralized, and scalable nature. However, it is vulnerable to poisoning attacks, and existing defenses either incur prohibitive resource overheads or rely on domain-specific prior knowledge, limiting their practical deployment. To address these limitations, we propose PEEL, a Poisoning-Exposing Encoding theoretical framework for LDP, which departs from resource- or prior-dependent countermeasures and instead leverages the inherent structural consistency of LDP-perturbed data. As a non-intrusive post-processing module, PEEL amplifies stealthy poisoning effects by re-encoding LDP-perturbed data via sparsification, normalization, and low-rank projection, thereby revealing both output and rule poisoning attacks through structural inconsistencies in the reconstructed space. Theoretical analysis proves that PEEL, integrated with LDP, retains unbiasedness and statistical accuracy, while being robust to expose both output and rule poisoning attacks. Moreover, evaluation results show that LDP-integrated PEEL not only outperforms four state-of-the-art defenses in terms of poisoning exposure accuracy but also significantly reduces client-side computational costs, making it highly suitable for large-scale IoT deployments.

cs.CR↗

arXiv2025

Text-Based Approaches to Item Alignment to Content Standards in Large-Scale Reading & Writing Tests

Aligning test items to content standards is a critical step in test development to collect validity evidence based on content. Item alignment has typically been conducted by human experts. This judgmental process can be subjective and time-consuming. This study investigated the performance of fine-tuned small language models (SLMs) for automated item alignment using data from a large-scale standardized reading and writing test for college admissions. Different SLMs were trained for alignment at both domain and skill levels respectively with 10 skills mapped to 4 content domains. The model performance was evaluated in multiple criteria on two testing datasets. The impact of types and sizes of the input data for training was investigated. Results showed that including more item text data led to substantially better model performance, surpassing the improvements induced by sample size increase alone. For comparison, supervised machine learning models were trained using the embeddings from the multilingual-E5-large-instruct model. The study results showed that fine-tuned SLMs consistently outperformed the embedding-based supervised machine learning models, particularly for the more fine-grained skill alignment. To better understand model misclassifications, multiple semantic similarity analysis including pairwise cosine similarity, Kullback-Leibler divergence of embedding distributions, and two-dimension projections of item embeddings were conducted. These analyses consistently showed that certain skills in SAT and PSAT were semantically too close, providing evidence for the observed misclassification.

cs.CL↗

arXiv2025

Automated Alignment of Math Items to Content Standards in Large-Scale Assessments Using Language Models

Accurate alignment of items to content standards is critical for valid score interpretation in large-scale assessments. This study evaluates three automated paradigms for aligning items with four domain and nineteen skill labels. First, we extracted embeddings and trained multiple classical supervised machine learning models, and further investigated the impact of dimensionality reduction on model performance. Second, we fine-tuned eight BERT model and its variants for both domain and skill alignment. Third, we explored ensemble learning with majority voting and stacking with multiple meta-models. The DeBERTa-v3-base achieved the highest weighted-average F1 score of 0.950 for domain alignment while the RoBERTa-large yielded the highest F1 score of 0.869 for skill alignment. Ensemble models did not surpass the best-performing language models. Dimension reduction enhanced linear classifiers based on embeddings but did not perform better than language models. This study demonstrated different methods in automated item alignment to content standards.}

cs.CL↗

arXiv2025

Continuous Subgraph Matching via Cost-Model-based Dynamic Vertex Dominance Embeddings (Technical Report)

In many real-world applications such as social network analysis, knowledge graph discovery, biological network analytics, and so on, graph data management has become increasingly important and has drawn much attention from the database community. While many graphs (e.g., Twitter, Wikipedia, etc.) are usually evolving over time, it is of great importance to study the \textit{continuous subgraph matching} (CSM) problem, a fundamental, yet challenging, graph operator, which continuously monitors subgraph matching results over dynamic graphs with a stream of edge updates. To efficiently tackle the CSM problem, we carefully design a general CSM processing framework, based on novel \textit{\underline{D}ynam\underline{I}c \underline{V}ertex Dom\underline{IN}ance \underline{E}mbedding} (DIVINE), which maps vertex neighborhoods into an embedding space to enable efficient subgraph matching and incremental maintenance under dynamic updates. Inspired by low pruning power for high-degree vertices, we propose a new \textit{degree grouping} technique to decompose high-degree star patterns into groups of lower-degree star substructures, and devise \textit{degree-aware star substructure synopses} (DAS$^3$) over embeddings of star substructure groups. We develop efficient algorithms to incrementally maintain dynamic graphs and answer CSM queries by traversing DAS$^3$ synopses and applying our designed \textit{vertex dominance} and \textit{range pruning strategies}. Through extensive experiments, we confirm the efficiency of our proposed DIVINE approach over both real and synthetic graphs.

cs.DB↗

arXiv2025

Text-Based Approaches to Item Difficulty Modeling in Large-Scale Assessments: A Systematic Review

Item difficulty plays a crucial role in test performance, interpretability of scores, and equity for all test-takers, especially in large-scale assessments. Traditional approaches to item difficulty modeling rely on field testing and classical test theory (CTT)-based item analysis or item response theory (IRT) calibration, which can be time-consuming and costly. To overcome these challenges, text-based approaches leveraging machine learning and language models, have emerged as promising alternatives. This paper reviews and synthesizes 37 articles on automated item difficulty prediction in large-scale assessment settings published through May 2025. For each study, we delineate the dataset, difficulty parameter, subject domain, item type, number of items, training and test data split, input, features, model, evaluation criteria, and model performance outcomes. Results showed that although classic machine learning models remain relevant due to their interpretability, state-of-the-art language models, using both small and large transformer-based architectures, can capture syntactic and semantic patterns without the need for manual feature engineering. Uniquely, model performance outcomes were summarized to serve as a benchmark for future research and overall, text-based methods have the potential to predict item difficulty with root mean square error (RMSE) as low as 0.165, Pearson correlation as high as 0.87, and accuracy as high as 0.806. The review concludes by discussing implications for practice and outlining future research directions for automated item difficulty modeling.

cs.CL↗

arXiv2025

A Chain-of-thought Reasoning Breast Ultrasound Dataset Covering All Histopathology Categories

Breast ultrasound (BUS) is an essential tool for diagnosing breast lesions, with millions of examinations per year. However, publicly available high-quality BUS benchmarks for AI development are limited in data scale and annotation richness. In this work, we present BUS-CoT, a BUS dataset for chain-of-thought (CoT) reasoning analysis, which contains 11,439 images of 10,019 lesions from 4,838 patients and covers all 99 histopathology types. To facilitate research on incentivizing CoT reasoning, we construct the reasoning processes based on observation, feature, diagnosis and pathology labels, annotated and verified by experienced experts. Moreover, by covering lesions of all histopathology types, we aim to facilitate robust AI systems in rare cases, which can be error-prone in clinical practice.

eess.IV↗

arXiv2025

Understanding the Thinking Process of Reasoning Models: A Perspective from Schoenfeld's Episode Theory

While Large Reasoning Models (LRMs) generate extensive chain-of-thought reasoning, we lack a principled framework for understanding how these thoughts are structured. In this paper, we introduce a novel approach by applying Schoenfeld's Episode Theory, a classic cognitive framework for human mathematical problem-solving, to analyze the reasoning traces of LRMs. We annotated thousands of sentences and paragraphs from model-generated solutions to math problems using seven cognitive labels (e.g., Plan, Implement, Verify). The result is the first publicly available benchmark for the fine-grained analysis of machine reasoning, including a large annotated corpus and detailed annotation guidebooks. Our preliminary analysis reveals distinct patterns in LRM reasoning, such as the transition dynamics between cognitive states. This framework provides a theoretically grounded methodology for interpreting LRM cognition and enables future work on more controllable and transparent reasoning systems.

cs.AI↗

arXiv2025

MIND: Towards Immersive Psychological Healing with Multi-agent Inner Dialogue

Mental health issues are worsening in today's competitive society, such as depression and anxiety. Traditional healings like counseling and chatbots fail to engage effectively, they often provide generic responses lacking emotional depth. Although large language models (LLMs) have the potential to create more human-like interactions, they still struggle to capture subtle emotions. This requires LLMs to be equipped with human-like adaptability and warmth. To fill this gap, we propose the MIND (Multi-agent INner Dialogue), a novel paradigm that provides more immersive psychological healing environments. Considering the strong generative and role-playing ability of LLM agents, we predefine an interactive healing framework and assign LLM agents different roles within the framework to engage in interactive inner dialogues with users, thereby providing an immersive healing experience. We conduct extensive human experiments in various real-world healing dimensions, and find that MIND provides a more user-friendly experience than traditional paradigms. This demonstrates that MIND effectively leverages the significant potential of LLMs in psychological healing.

cs.CL↗