arXiv subjects

Yuchen Zhang

Publications and source records attributed to Yuchen Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2025

A Deep Learning Framework for Joint Channel Acquisition and Communication Optimization in Movable Antenna Systems

This paper presents an end-to-end deep learning framework in a movable antenna (MA)-enabled multiuser communication system. In contrast to the conventional works assuming perfect channel state information (CSI), we address the practical CSI acquisition issue through the design of pilot signals and quantized CSI feedback, and further incorporate the joint optimization of channel estimation, MA placement, and precoding design. The proposed mechanism enables the system to learn an optimized transmission strategy from imperfect channel data, overcoming the limitations of conventional methods that conduct channel estimation and antenna position optimization separately. To balance the performance and overhead, we further extend the proposed framework to optimize the antenna placement based on the statistical CSI. Simulation results demonstrate that the proposed approach consistently outperforms traditional benchmarks in terms of achievable sum-rate of users, especially under limited feedback and sparse channel environments. Notably, it achieves a performance comparable to the widely-adopted gradient-based methods with perfect CSI, while maintaining significantly lower CSI feedback overhead. These results highlight the effectiveness and adaptability of learning-based MA system design for future wireless systems.

cs.IT↗

arXiv2025

Hybrid Codebook Design for Localization Using Electromagnetically Reconfigurable Fluid Antenna System

Electromagnetically reconfigurable fluid antenna systems (ER-FAS) introduce additional degrees of freedom in the electromagnetic (EM) domain by dynamically steering per-antenna radiation patterns, thereby enhancing power efficiency in wireless links. Unlike prior works on spatially reconfigurable FAS, which adjust element positions, ER-FAS provides direct control over each element's EM characteristics to realize on-demand beam-pattern shaping. While existing studies have exploited ER-FAS to boost spectral efficiency, this paper explores its application for downlink localization. We consider a multiple-input single-output (MISO) system in which a multi-antenna ER-FAS at the base station serves a single-antenna user equipment (UE). We consider two reconfigurability paradigms: (i) a synthesis model where each antenna generates desired beampatterns from a finite set of EM basis functions, and (ii) a finite-state selection model in which each antenna selects a pattern from a predefined set of patterns. For both paradigms, we formulate the joint baseband (BB) and EM precoder design to minimize the UE position error bound. In the synthesis case we derive low-dimensional closed-form expressions for both the BB and EM precoders. For the finite-state model we obtain closed-form BB structures and propose a low-complexity block-coordinate-descent algorithm for EM pattern selection. Analytical bounds and extensive simulations show that the proposed hybrid designs for ER-FAS substantially improve UE positioning accuracy over traditional non-reconfigurable arrays.

eess.SP↗

arXiv2025

Symbolic Equation Modeling of Composite Loads: A Kolmogorov-Arnold Network based Learning Approach

With increasing penetration of distributed energy resources installed behind the meter, there is a growing need for adequate modelling of composite loads to enable accurate power system simulation analysis. Existing measurement based load modeling methods either fit fixed-structure physical models, which limits adaptability to evolving load mixes, or employ flexible machine learning methods which are however black boxes and offer limited interpretability. This paper presents a new learning based load modelling method based on Kolmogorov Arnold Networks towards modelling flexibility and interpretability. By actively learning activation functions on edges, KANs automatically derive free form symbolic equations that capture nonlinear relationships among measured variables without prior assumptions about load structure. Case studies demonstrate that the proposed approach outperforms other methods in both accuracy and generalization ability, while uniquely representing composite loads into transparent, interpretable mathematical equations.

eess.SY↗

arXiv2025

SSRL: Self-Search Reinforcement Learning

We investigate the potential of large language models (LLMs) to serve as efficient simulators for agentic search tasks in reinforcement learning (RL), thereby reducing dependence on costly interactions with external search engines. To this end, we first quantify the intrinsic search capability of LLMs via structured prompting and repeated sampling, which we term Self-Search. Our results reveal that LLMs exhibit strong scaling behavior with respect to the inference budget, achieving high pass@k on question-answering benchmarks, including the challenging BrowseComp task. Building on these observations, we introduce Self-Search RL (SSRL), which enhances LLMs' Self-Search capability through format-based and rule-based rewards. SSRL enables models to iteratively refine their knowledge utilization internally, without requiring access to external tools. Empirical evaluations demonstrate that SSRL-trained policy models provide a cost-effective and stable environment for search-driven RL training, reducing reliance on external search engines and facilitating robust sim-to-real transfer. We draw the following conclusions: 1) LLMs possess world knowledge that can be effectively elicited to achieve high performance; 2) SSRL demonstrates the potential of leveraging internal knowledge to reduce hallucination; 3) SSRL-trained models integrate seamlessly with external search engines without additional effort. Our findings highlight the potential of LLMs to support more scalable RL agent training.

cs.CL↗

arXiv2025

Echo: Decoupling Inference and Training for Large-Scale RL Alignment on Heterogeneous Swarms

Modern RL-based post-training for large language models (LLMs) co-locate trajectory sampling and policy optimisation on the same GPU cluster, forcing the system to switch between inference and training workloads. This serial context switching violates the single-program-multiple-data (SPMD) assumption underlying today's distributed training systems. We present Echo, the RL system that cleanly decouples these two phases across heterogeneous "inference" and "training" swarms while preserving statistical efficiency. Echo introduces two lightweight synchronization protocols: a sequential pull mode that refreshes policy weights according to API call for minimal bias, and an asynchronous push-pull mode that streams version-tagged rollouts through a replay buffer to maximise hardware utilisation. Training four representative RL workloads with Qwen3-4B, Qwen2.5-7B, Qwen3-30B-A3B-Thinking-2507 and Qwen3-32B on a geographically distributed cluster, Echo matches a fully co-located Verl baseline in convergence speed and final reward while off-loading trajectory generation to commodity edge hardware. These promising results demonstrate that large-scale RL for LLMs could achieve datacentre-grade performance using decentralised, heterogeneous resources.

cs.LG↗

arXiv2025

StoryTeller: Improving Long Video Description through Global Audio-Visual Character Identification

Existing large vision-language models (LVLMs) are largely limited to processing short, seconds-long videos and struggle with generating coherent descriptions for extended video spanning minutes or more. Long video description introduces new challenges, such as consistent character identification and plot-level descriptions incorporating both visual and audio information. To address these, we figure out audio-visual character identification, matching character names to each dialogue, as a key factor. We propose StoryTeller, a system for generating dense descriptions of long videos, incorporating both low-level visual concepts and high-level plot information. StoryTeller uses a multimodal large language model that integrates visual, audio, and text modalities to perform audio-visual character identification on minute-long video clips. The results are then fed into a LVLM to enhance consistency of video description. We validate our approach on movie description tasks and introduce MovieStory101, a dataset with dense descriptions for three-minute movie clips. To evaluate long video descriptions, we create StoryQA, a large set of multiple-choice questions for MovieStory101 test set. We assess descriptions by inputting them into GPT-4 to answer these questions, using accuracy as an automatic evaluation metric. Experiments show that StoryTeller outperforms all open and closed-source baselines on StoryQA, achieving 9.5% higher accuracy than the strongest baseline, Gemini-1.5-pro, and demonstrating a +15.56% advantage in human side-by-side evaluations. Additionally, incorporating audio-visual character identification from StoryTeller improves the performance of all video description models, with Gemini-1.5-pro and GPT-4o showing relative improvement of 5.5% and 13.0%, respectively, in accuracy on StoryQA.

cs.CV↗

arXiv2025

TTRL: Test-Time Reinforcement Learning

This paper investigates Reinforcement Learning (RL) on data without explicit labels for reasoning tasks in Large Language Models (LLMs). The core challenge of the problem is reward estimation during inference while not having access to ground-truth information. While this setting appears elusive, we find that common practices in Test-Time Scaling (TTS), such as majority voting, yield surprisingly effective rewards suitable for driving RL training. In this work, we introduce Test-Time Reinforcement Learning (TTRL), a novel method for training LLMs using RL on unlabeled data. TTRL enables self-evolution of LLMs by utilizing the priors in the pre-trained models. Our experiments demonstrate that TTRL consistently improves performance across a variety of tasks and models. Notably, TTRL boosts the pass@1 performance of Qwen-2.5-Math-7B by approximately 211% on the AIME 2024 with only unlabeled test data. Furthermore, although TTRL is only supervised by the maj@n metric, TTRL has demonstrated performance to consistently surpass the upper limit of the initial model maj@n, and approach the performance of models trained directly on test data with ground-truth labels. Our experimental findings validate the general effectiveness of TTRL across various tasks and highlight TTRL's potential for broader tasks and domains. GitHub: https://github.com/PRIME-RL/TTRL

cs.CL↗

arXiv2025

Unlearning-Enhanced Website Fingerprinting Attack: Against Backdoor Poisoning in Anonymous Networks

Website Fingerprinting (WF) is an effective tool for regulating and governing the dark web. However, its performance can be significantly degraded by backdoor poisoning attacks in practical deployments. This paper aims to address the problem of hidden backdoor poisoning attacks faced by Website Fingerprinting attack, and designs a feasible mothed that integrates unlearning technology to realize detection of automatic poisoned points and complete removal of its destructive effects, requiring only a small number of known poisoned test points. Taking Tor onion routing as an example, our method evaluates the influence value of each training sample on these known poisoned test points as the basis for judgment. We optimize the use of influence scores to identify poisoned samples within the training dataset. Furthermore, by quantifying the difference between the contribution of model parameters on the taining data and the clean data, the target parameters are dynamically adjusted to eliminate the impact of the backdoor attacks. Experiments on public datasets under the assumptions of closed-world (CW) and open-world (OW) verify the effectiveness of the proposed method. In complex scenes containing both clean website fingerprinting features and backdoor triggers, the accuracy of the model on the poisoned dataset and the test dataset is stable at about 80%, significantly outperforming the traditional WF attack models. In addition, the proposed method achieves a 2-3 times speedup in runtime efficiency compared to baseline methods. By incorporating machine unlearning, we realize a WF attack model that exhibits enhanced resistance to backdoor poisoning and faster execution speeds in adversarial settings.

cs.CR↗

arXiv2025

xbench: Tracking Agents Productivity Scaling with Profession-Aligned Real-World Evaluations

We introduce xbench, a dynamic, profession-aligned evaluation suite designed to bridge the gap between AI agent capabilities and real-world productivity. While existing benchmarks often focus on isolated technical skills, they may not accurately reflect the economic value agents deliver in professional settings. To address this, xbench targets commercially significant domains with evaluation tasks defined by industry professionals. Our framework creates metrics that strongly correlate with productivity value, enables prediction of Technology-Market Fit (TMF), and facilitates tracking of product capabilities over time. As our initial implementations, we present two benchmarks: Recruitment and Marketing. For Recruitment, we collect 50 tasks from real-world headhunting business scenarios to evaluate agents' abilities in company mapping, information retrieval, and talent sourcing. For Marketing, we assess agents' ability to match influencers with advertiser needs, evaluating their performance across 50 advertiser requirements using a curated pool of 836 candidate influencers. We present initial evaluation results for leading contemporary agents, establishing a baseline for these professional domains. Our continuously updated evalsets and evaluations are available at https://xbench.org.

cs.LG↗

arXiv2025

The Entropy Mechanism of Reinforcement Learning for Reasoning Language Models

This paper aims to overcome a major obstacle in scaling RL for reasoning with LLMs, namely the collapse of policy entropy. Such phenomenon is consistently observed across vast RL runs without entropy intervention, where the policy entropy dropped sharply at the early training stage, this diminished exploratory ability is always accompanied with the saturation of policy performance. In practice, we establish a transformation equation R=-a*e^H+b between entropy H and downstream performance R. This empirical law strongly indicates that, the policy performance is traded from policy entropy, thus bottlenecked by its exhaustion, and the ceiling is fully predictable H=0, R=-a+b. Our finding necessitates entropy management for continuous exploration toward scaling compute for RL. To this end, we investigate entropy dynamics both theoretically and empirically. Our derivation highlights that, the change in policy entropy is driven by the covariance between action probability and the change in logits, which is proportional to its advantage when using Policy Gradient-like algorithms. Empirical study shows that, the values of covariance term and entropy differences matched exactly, supporting the theoretical conclusion. Moreover, the covariance term stays mostly positive throughout training, further explaining why policy entropy would decrease monotonically. Through understanding the mechanism behind entropy dynamics, we motivate to control entropy by restricting the update of high-covariance tokens. Specifically, we propose two simple yet effective techniques, namely Clip-Cov and KL-Cov, which clip and apply KL penalty to tokens with high covariances respectively. Experiments show that these methods encourage exploration, thus helping policy escape entropy collapse and achieve better downstream performance.

cs.LG↗

arXiv2025

PaSa: An LLM Agent for Comprehensive Academic Paper Search

We introduce PaSa, an advanced Paper Search agent powered by large language models. PaSa can autonomously make a series of decisions, including invoking search tools, reading papers, and selecting relevant references, to ultimately obtain comprehensive and accurate results for complex scholar queries. We optimize PaSa using reinforcement learning with a synthetic dataset, AutoScholarQuery, which includes 35k fine-grained academic queries and corresponding papers sourced from top-tier AI conference publications. Additionally, we develop RealScholarQuery, a benchmark collecting real-world academic queries to assess PaSa performance in more realistic scenarios. Despite being trained on synthetic data, PaSa significantly outperforms existing baselines on RealScholarQuery, including Google, Google Scholar, Google with GPT-4o for paraphrased queries, ChatGPT (search-enabled GPT-4o), GPT-o1, and PaSa-GPT-4o (PaSa implemented by prompting GPT-4o). Notably, PaSa-7B surpasses the best Google-based baseline, Google with GPT-4o, by 37.78% in recall@20 and 39.90% in recall@50, and exceeds PaSa-GPT-4o by 30.36% in recall and 4.25% in precision. Model, datasets, and code are available at https://github.com/bytedance/pasa.

cs.IR↗

arXiv2025

Enhanced Multimodal Hate Video Detection via Channel-wise and Modality-wise Fusion

The rapid rise of video content on platforms such as TikTok and YouTube has transformed information dissemination, but it has also facilitated the spread of harmful content, particularly hate videos. Despite significant efforts to combat hate speech, detecting these videos remains challenging due to their often implicit nature. Current detection methods primarily rely on unimodal approaches, which inadequately capture the complementary features across different modalities. While multimodal techniques offer a broader perspective, many fail to effectively integrate temporal dynamics and modality-wise interactions essential for identifying nuanced hate content. In this paper, we present CMFusion, an enhanced multimodal hate video detection model utilizing a novel Channel-wise and Modality-wise Fusion Mechanism. CMFusion first extracts features from text, audio, and video modalities using pre-trained models and then incorporates a temporal cross-attention mechanism to capture dependencies between video and audio streams. The learned features are then processed by channel-wise and modality-wise fusion modules to obtain informative representations of videos. Our extensive experiments on a real-world dataset demonstrate that CMFusion significantly outperforms five widely used baselines in terms of accuracy, precision, recall, and F1 score. Comprehensive ablation studies and parameter analyses further validate our design choices, highlighting the model's effectiveness in detecting hate videos. The source codes will be made publicly available at https://github.com/EvelynZ10/cmfusion.

cs.MM↗

arXiv2025

Tri-Hybrid Multi-User Precoding Based on Electromagnetically Reconfigurable Antennas

The tri-hybrid precoding architecture based on electromagnetically reconfigurable antennas (ERAs) is a promising solution for overcoming key limitations in multiple-input multiple-output communication systems. Aiming to further understand its potential, this paper investigates the tri-hybrid multi-user precoding problem using pattern reconfigurable ERAs. To reduce model complexity and improve practicality, we characterize each antenna's radiation pattern using a spherical harmonics decomposition. While mathematically tractable, this approach may lead to over-optimized patterns that are physically unrealizable. To address this, we introduce a projection step that maps the optimized patterns onto a realizable set. Simulation results demonstrate that spherical harmonics-based radiation pattern optimization significantly enhances sum rate performance. However, after projection onto a realizable set obtained from real ERA hardware, the performance gain is notably reduced or even negligible, underscoring the need for more effective projection techniques and improved reconfigurable antenna hardware.

eess.SP↗

arXiv2025

Association between nutritional factors, inflammatory biomarkers and cancer types: an analysis of NHANES data using machine learning

Background. Diet and inflammation are critical factors influencing cancer risk. However, the combined impact of nutritional status and inflammatory biomarkers on cancer status and type, using machine learning (ML), remains underexplored. Objectives. This study investigates the association between nutritional factors, inflammatory biomarkers, and cancer status, and whether these relationships differ across cancer types using National Health and Nutrition Examination Survey (NHANES) data. Methods. We analyzed 24 macro- and micronutrients, C-reactive protein (CRP), and the advanced lung cancer inflammation index (ALI) in 26,409 NHANES participants (2,120 with cancer). Multivariable logistic regression assessed associations with cancer prevalence. We also examined whether these features differed across the five most common cancer types. To evaluate predictive value, we applied three ML models - Logistic Regression, Random Forest, and XGBoost - on the full feature set. Results. The cohort's mean age was 49.1 years; 34.7% were obese. Comorbidities such as anemia and liver conditions, along with nutritional factors like protein and several vitamins, were key predictors of cancer status. Among the models, Random Forest performed best, achieving an accuracy of 0.72. Conclusions. Higher-quality nutritional intake and lower levels of inflammation may offer protective effects against cancer. These findings highlight the potential of combining nutritional and inflammatory markers with ML to inform cancer prevention strategies.

q-bio.QM↗

arXiv2025

Cross-Frequency Collaborative Training Network and Dataset for Semi-supervised First Molar Root Canal Segmentation

Root canal (RC) treatment is a highly delicate and technically complex procedure in clinical practice, heavily influenced by the clinicians' experience and subjective judgment. Deep learning has made significant advancements in the field of computer-aided diagnosis (CAD) because it can provide more objective and accurate diagnostic results. However, its application in RC treatment is still relatively rare, mainly due to the lack of public datasets in this field. To address this issue, in this paper, we established a First Molar Root Canal segmentation dataset called FMRC-2025. Additionally, to alleviate the workload of manual annotation for dentists and fully leverage the unlabeled data, we designed a Cross-Frequency Collaborative training semi-supervised learning (SSL) Network called CFC-Net. It consists of two components: (1) Cross-Frequency Collaborative Mean Teacher (CFC-MT), which introduces two specialized students (SS) and one comprehensive teacher (CT) for collaborative multi-frequency training. The CT and SS are trained on different frequency components while fully integrating multi-frequency knowledge through cross and full frequency consistency supervisions. (2) Uncertainty-guided Cross-Frequency Mix (UCF-Mix) mechanism enables the network to generate high-confidence pseudo-labels while learning to integrate multi-frequency information and maintaining the structural integrity of the targets. Extensive experiments on FMRC-2025 and three public dental datasets demonstrate that CFC-MT is effective for RC segmentation and can also exhibit strong generalizability on other dental segmentation tasks, outperforming state-of-the-art SSL medical image segmentation methods. Codes and dataset will be released.

cs.CV↗

arXiv2025

PRAD: Periapical Radiograph Analysis Dataset and Benchmark Model Development

Deep learning (DL), a pivotal technology in artificial intelligence, has recently gained substantial traction in the domain of dental auxiliary diagnosis. However, its application has predominantly been confined to imaging modalities such as panoramic radiographs and Cone Beam Computed Tomography, with limited focus on auxiliary analysis specifically targeting Periapical Radiographs (PR). PR are the most extensively utilized imaging modality in endodontics and periodontics due to their capability to capture detailed local lesions at a low cost. Nevertheless, challenges such as resolution limitations and artifacts complicate the annotation and recognition of PR, leading to a scarcity of publicly available, large-scale, high-quality PR analysis datasets. This scarcity has somewhat impeded the advancement of DL applications in PR analysis. In this paper, we present PRAD-10K, a dataset for PR analysis. PRAD-10K comprises 10,000 clinical periapical radiograph images, with pixel-level annotations provided by professional dentists for nine distinct anatomical structures, lesions, and artificial restorations or medical devices, We also include classification labels for images with typical conditions or lesions. Furthermore, we introduce a DL network named PRNet to establish benchmarks for PR segmentation tasks. Experimental results demonstrate that PRNet surpasses previous state-of-the-art medical image segmentation models on the PRAD-10K dataset. The codes and dataset will be made publicly available.

eess.IV↗

arXiv2025

Adversarial Curriculum Graph-Free Knowledge Distillation for Graph Neural Networks

Data-free Knowledge Distillation (DFKD) is a method that constructs pseudo-samples using a generator without real data, and transfers knowledge from a teacher model to a student by enforcing the student to overcome dimensional differences and learn to mimic the teacher's outputs on these pseudo-samples. In recent years, various studies in the vision domain have made notable advancements in this area. However, the varying topological structures and non-grid nature of graph data render the methods from the vision domain ineffective. Building upon prior research into differentiable methods for graph neural networks, we propose a fast and high-quality data-free knowledge distillation approach in this paper. Without compromising distillation quality, the proposed graph-free KD method (ACGKD) significantly reduces the spatial complexity of pseudo-graphs by leveraging the Binary Concrete distribution to model the graph structure and introducing a spatial complexity tuning parameter. This approach enables efficient gradient computation for the graph structure, thereby accelerating the overall distillation process. Additionally, ACGKD eliminates the dimensional ambiguity between the student and teacher models by increasing the student's dimensions and reusing the teacher's classifier. Moreover, it equips graph knowledge distillation with a CL-based strategy to ensure the student learns graph structures progressively. Extensive experiments demonstrate that ACGKD achieves state-of-the-art performance in distilling knowledge from GNNs without training data.

cs.LG↗

arXiv2025

Semantic SLAM with Rolling-Shutter Cameras and Low-Precision INS in Outdoor Environments

Accurate localization and mapping in outdoor environments remains challenging when using consumer-grade hardware, particularly with rolling-shutter cameras and low-precision inertial navigation systems (INS). We present a novel semantic SLAM approach that leverages road elements such as lane boundaries, traffic signs, and road markings to enhance localization accuracy. Our system integrates real-time semantic feature detection with a graph optimization framework, effectively handling both rolling-shutter effects and INS drift. Using a practical hardware setup which consists of a rolling-shutter camera (3840*2160@30fps), IMU (100Hz), and wheel encoder (50Hz), we demonstrate significant improvements over existing methods. Compared to state-of-the-art approaches, our method achieves higher recall (up to 5.35\%) and precision (up to 2.79\%) in semantic element detection, while maintaining mean relative error (MRE) within 10cm and mean absolute error (MAE) around 1m. Extensive experiments in diverse urban environments demonstrate the robust performance of our system under varying lighting conditions and complex traffic scenarios, making it particularly suitable for autonomous driving applications. The proposed approach provides a practical solution for high-precision localization using affordable hardware, bridging the gap between consumer-grade sensors and production-level performance requirements.

cs.RO↗