arXiv subjects

Zheng Gao

Publications and source records attributed to Zheng Gao.

At least 37 records · Page 2Linked to original sources

arXiv2025

BondBERT: What we learn when assigning sentiment in the bond market

Bond markets respond differently to macroeconomic news compared to equity markets, yet most sentiment models are trained primarily on general financial or equity news data. However, bond prices often move in the opposite direction to economic optimism, making general or equity-based sentiment tools potentially misleading. We introduce BondBERT, a transformer-based language model fine-tuned on bond-specific news. BondBERT can act as the perception and reasoning component of a financial decision-support agent, providing sentiment signals that integrate with forecasting models. We propose a generalisable framework for adapting transformers to low-volatility, domain-inverse sentiment tasks by compiling and cleaning 30,000 UK bond market articles (2018-2025). BondBERT's sentiment predictions are compared against FinBERT, FinGPT, and Instruct-FinGPT using event-based correlation, up/down accuracy analyses, and LSTM forecasting across ten UK sovereign bonds. We find that BondBERT consistently produces positive correlations with bond returns, and achieves higher alignment and forecasting accuracy than the three baseline models. These results demonstrate that domain-specific sentiment adaptation better captures fixed income dynamics, bridging a gap between NLP advances and bond market analytics.

q-fin.CP↗

arXiv2025

Robust and Hyper-Efficient Multi-dimensional Optical Fiber Semantic Communication

The growing demands of artificial intelligence and immersive media require communication beyond bit-level accuracy to meaning awareness. Conventional optical systems that focused on syntactic precision suffer significant inefficiencies. Here, we introduce a multi-dimensional semantic communication framework that bridges this gap by directly mapping high-level semantic features onto the orthogonal physical dimensions of light, frequency, polarization, and intensity, within a multimode fiber. This synergistic co-design of semantic logic and the photonic channel achieve an unprecedented equivalent spectral efficiency approaching 1000 bit/s/Hz. Moreover, it demonstrates profound resilience, maintaining high-fidelity reconstruction even when the physical-layer symbol error rate exceeds 36%, a condition under which conventional communication systems fail completely. Crucially, this deeply integrated co-design of semantic encoding and physical-layer modulation enables full semantic demodulation with only single-ended intensity detection, therefore significantly reducing system complexity and cost. This work establishes a validated pathway toward hyper-efficient, error-resilient optical networks for the next generation of data-intensive computing.

physics.optics↗

arXiv2025

Are LLMs Better GNN Helpers? Rethinking Robust Graph Learning under Deficiencies with Iterative Refinement

Graph Neural Networks (GNNs) are widely adopted in Web-related applications, serving as a core technique for learning from graph-structured data, such as text-attributed graphs. Yet in real-world scenarios, such graphs exhibit deficiencies that substantially undermine GNN performance. While prior GNN-based augmentation studies have explored robustness against individual imperfections, a systematic understanding of how graph-native and Large Language Models (LLMs) enhanced methods behave under compound deficiencies is still missing. Specifically, there has been no comprehensive investigation comparing conventional approaches and recent LLM-on-graph frameworks, leaving their merits unclear. To fill this gap, we conduct the first empirical study that benchmarks these two lines of methods across diverse graph deficiencies, revealing overlooked vulnerabilities and challenging the assumption that LLM augmentation is consistently superior. Building on empirical findings, we propose Robust Graph Learning via Retrieval-Augmented Contrastive Refinement (RoGRAD) framework. Unlike prior one-shot LLM-as-Enhancer designs, RoGRAD is the first iterative paradigm that leverages Retrieval-Augmented Generation (RAG) to inject retrieval-grounded augmentations by supplying class-consistent, diverse augmentations and enforcing discriminative representations through iterative graph contrastive learning. It transforms LLM augmentation for graphs from static signal injection into dynamic refinement. Extensive experiments demonstrate RoGRAD's superiority over both conventional GNN- and LLM-enhanced baselines, achieving up to 82.43% average improvement.

cs.LG↗

arXiv2025

Science Out of Its Ivory Tower: Improving Accessibility with Reinforcement Learning

A vast amount of scholarly work is published daily, yet much of it remains inaccessible to the general public due to dense jargon and complex language. To address this challenge in science communication, we introduce a reinforcement learning framework that fine-tunes a language model to rewrite scholarly abstracts into more comprehensible versions. Guided by a carefully balanced combination of word- and sentence-level accessibility rewards, our language model effectively substitutes technical terms with more accessible alternatives, a task which models supervised fine-tuned or guided by conventional readability measures struggle to accomplish. Our best model adjusts the readability level of scholarly abstracts by approximately six U.S. grade levels -- in other words, from a postgraduate to a high school level. This translates to roughly a 90% relative boost over the supervised fine-tuning baseline, all while maintaining factual accuracy and high-quality language. An in-depth analysis of our approach shows that balanced rewards lead to systematic modifications in the base model, likely contributing to smoother optimization and superior performance. We envision this work as a step toward bridging the gap between scholarly research and the general public, particularly younger readers and those without a college degree.

cs.CL↗

arXiv2025

Citation Recommendation based on Argumentative Zoning of User Queries

Citation recommendation aims to locate the important papers for scholars to cite. When writing the citing sentences, the authors usually hold different citing intents, which are referred to citation function in citation analysis. Since argumentative zoning is to identify the argumentative and rhetorical structure in scientific literature, we want to use this information to improve the citation recommendation task. In this paper, a multi-task learning model is built for citation recommendation and argumentative zoning classification. We also generated an annotated corpus of the data from PubMed Central based on a new argumentative zoning schema. The experimental results show that, by considering the argumentative information in the citing sentence, citation recommendation model will get better performance.

cs.IR↗

arXiv2024

Optical Semantic Communication through Multimode Fiber: From Symbol Transmission to Sentiment Analysis

We propose and validate a novel optical semantic transmission scheme using multimode fiber (MMF). By leveraging the frequency sensitivity of intermodal dispersion in MMFs, we achieve high-dimensional semantic encoding and decoding in the frequency domain. Our system maps symbols to 128 distinct frequencies spaced at 600 kHz intervals, demonstrating a seven-fold increase in capacity compared to conventional communication encoding. We further enhance spectral efficiency by implementing 4-level pulse amplitude modulation (PAM-4), achieving 9.12 bits/s/Hz without decoding errors. Additionally, we explore the application of this system for sentiment analysis using the IMDb movie review dataset. By encoding semantically similar symbols to adjacent frequencies, the system's noise tolerance is effectively improved, facilitating accurate sentiment analysis. This work highlights the potential of MMF-based semantic communication to enhance both capacity and robustness in optical communication systems, offering promising applications in bandwidth-constrained and noisy environments.

eess.SP↗

arXiv2024

Stable Acceleration of a LHe-Free Nb3Sn demo SRF e-linac Based on Conduction Cooling

The design, construction, and commissioning of a conduction-cooled Nb3Sn demonstration superconducting radio frequency (SRF) electron accelerator at the Institute of Modern Physics of the Chinese Academy of Sciences (IMP, CAS) will be presented. In the context of engineering application planning for Nb3Sn thin-film SRF cavities within the CiADS project, a 650MHz 5-cell elliptical cavity was coated using the vapor diffusion method for electron beam acceleration. Through high-precision collaborative control of 10 GM cryocooler, slow cooldown of the cavity crossing 18K is achieved accompanied by obviously characteristic magnetic flux expulsion. The horizontal test results of the liquid helium-free (LHe-free) cryomodule show that the cavity can operate steadily at Epk=6.02MV/m in continuous wave (CW) mode, and at Epk=14.90MV/m in 40% duty cycle pulse mode. The beam acceleration experiment indicates that the maximum average current of the electron beam in the macropulse after acceleration exceeds 200mA, with a maximum energy gain of 4.6MeV. The results provide a principle validation for the engineering application of Nb3Sn thin-film SRF cavities, highlighting the promising industrial application prospects of a small-scale compact Nb3Sn SRF accelerator driven by commercial cryocoolers.

physics.acc-ph↗

arXiv2024

Self-Supervised Facial Representation Learning with Facial Region Awareness

Self-supervised pre-training has been proved to be effective in learning transferable representations that benefit various visual tasks. This paper asks this question: can self-supervised pre-training learn general facial representations for various facial analysis tasks? Recent efforts toward this goal are limited to treating each face image as a whole, i.e., learning consistent facial representations at the image-level, which overlooks the consistency of local facial representations (i.e., facial regions like eyes, nose, etc). In this work, we make a first attempt to propose a novel self-supervised facial representation learning framework to learn consistent global and local facial representations, Facial Region Awareness (FRA). Specifically, we explicitly enforce the consistency of facial regions by matching the local facial representations across views, which are extracted with learned heatmaps highlighting the facial regions. Inspired by the mask prediction in supervised semantic segmentation, we obtain the heatmaps via cosine similarity between the per-pixel projection of feature maps and facial mask embeddings computed from learnable positional embeddings, which leverage the attention mechanism to globally look up the facial image for facial regions. To learn such heatmaps, we formulate the learning of facial mask embeddings as a deep clustering problem by assigning the pixel features from the feature maps to them. The transfer learning results on facial classification and regression tasks show that our FRA outperforms previous pre-trained models and more importantly, using ResNet as the unified backbone for various tasks, our FRA achieves comparable or even better performance compared with SOTA methods in facial analysis tasks.

cs.CV↗

arXiv2024

Multimode fiber speckle Stokes polarimeter

The detection of the state of polarization (SOP) of light is essential for many optical applications. However, it is a challenge for cost-effective SOP measurement due to the complexity of conventional methods and poor transferability of new methods. Here, we propose a straightforward, low-cost and portable SOP measurement system based on the multimode fiber speckle. Convolutional neural network is utilized to establish the mapping relationship between speckle and Stokes parameters. The lowest root mean square error of the estimated SOP on Poincare sphere can be 0.0042. This method is distinguished by its low cost, clear structure and applicability to different wavelengths with high precision. The proposed method is of great value in polarization-related applications.

physics.optics↗

arXiv2023

Self-Supervised Representation Learning with Cross-Context Learning between Global and Hypercolumn Features

Whilst contrastive learning yields powerful representations by matching different augmented views of the same instance, it lacks the ability to capture the similarities between different instances. One popular way to address this limitation is by learning global features (after the global pooling) to capture inter-instance relationships based on knowledge distillation, where the global features of the teacher are used to guide the learning of the global features of the student. Inspired by cross-modality learning, we extend this existing framework that only learns from global features by encouraging the global features and intermediate layer features to learn from each other. This leads to our novel self-supervised framework: cross-context learning between global and hypercolumn features (CGH), that enforces the consistency of instance relations between low- and high-level semantics. Specifically, we stack the intermediate feature maps to construct a hypercolumn representation so that we can measure instance relations using two contexts (hypercolumn and global feature) separately, and then use the relations of one context to guide the learning of the other. This cross-context learning allows the model to learn from the differences between the two contexts. The experimental results on linear classification and downstream tasks show that our method outperforms the state-of-the-art methods.

cs.CV↗

arXiv2023

FrFT based estimation of linear and nonlinear impairments using Vision Transformer

To comprehensively assess optical fiber communication system conditions, it is essential to implement joint estimation of the following four critical impairments: nonlinear signal-to-noise ratio (SNRNL), optical signal-to-noise ratio (OSNR), chromatic dispersion (CD) and differential group delay (DGD). However, current studies only achieve identifying a limited number of impairments within a narrow range, due to limitations in network capabilities and lack of unified representation of impairments. To address these challenges, we adopt time-frequency signal processing based on fractional Fourier transform (FrFT) to achieve the unified representation of impairments, while employing a Transformer based neural networks (NN) to break through network performance limitations. To verify the effectiveness of the proposed estimation method, the numerical simulation is carried on a 5-channel polarization-division-multiplexed quadrature phase shift keying (PDM-QPSK) long haul optical transmission system with the symbol rate of 50 GBaud per channel, the mean absolute error (MAE) for SNRNL, OSNR, CD, and DGD estimation is 0.091 dB, 0.058 dB, 117 ps/nm, and 0.38 ps, and the monitoring window ranges from 0~20 dB, 10~30 dB, 0~51000 ps/nm, and 0~100 ps, respectively. Our proposed method achieves accurate estimation of linear and nonlinear impairments over a broad range, representing a significant advancement in the field of optical performance monitoring (OPM).

eess.SP↗

arXiv2023

Detection of Sparse Anomalies in High-Dimensional Network Telescope Signals

Network operators and system administrators are increasingly overwhelmed with incessant cyber-security threats ranging from malicious network reconnaissance to attacks such as distributed denial of service and data breaches. A large number of these attacks could be prevented if the network operators were better equipped with threat intelligence information that would allow them to block or throttle nefarious scanning activities. Network telescopes or "darknets" offer a unique window into observing Internet-wide scanners and other malicious entities, and they could offer early warning signals to operators that would be critical for infrastructure protection and/or attack mitigation. A network telescope consists of unused or "dark" IP spaces that serve no users, and solely passively observes any Internet traffic destined to the "telescope sensor" in an attempt to record ubiquitous network scanners, malware that forage for vulnerable devices, and other dubious activities. Hence, monitoring network telescopes for timely detection of coordinated and heavy scanning activities is an important, albeit challenging, task. The challenges mainly arise due to the non-stationarity and the dynamic nature of Internet traffic and, more importantly, the fact that one needs to monitor high-dimensional signals (e.g., all TCP/UDP ports) to search for "sparse" anomalies. We propose statistical methods to address both challenges in an efficient and "online" manner; our work is validated both with synthetic data as well as real-world data from a large network telescope.

cs.CR↗

arXiv2022

Dynamic structure factor of one-dimensional Fermi superfluid with spin-orbit coupling

We theoretically calculate the density dynamic structure factor of one-dimensional Fermi superfluid with Raman-type spin-orbit coupling, and analyze its main dynamical character during phase transition between Bardeen-Cooper-Schrieffer superfluid and topological superfluid. Our theoretical results display four kinds of single-particle excitations induced by the two-branch structure of single-particle spectrum, and the cross single-particle excitation is much easier to be seen in the spin dynamic structure factor at a small transferred momentum. Also we find a new roton-like collective mode emerges at a fixed transferred momentum $q \simeq 2k_F$, and it only appears once the system enters the topological superfluid state. The occurrence of this roton-like excitation is related to switch of global minimum in single-particle spectrum from $k=0$ to $k \simeq 2k_F$.

cond-mat.quant-gas↗

arXiv2022

Phase Transitions in Genome-wide Association Studies and Categorical Variable Screenings

Motivated by genome-wide association screening studies (GWAS), we study high-dimensional marginal screenings of categorical variables where test statistics have approximate chi-square distributions. We characterize four new phase transitions in high-dimensional chi-square models, and derive the signal sizes necessary and sufficient for statistical procedures to simultaneously control false discovery (in terms of family-wise error rate or false discovery rate) and missed detection (in terms of family-wise non-discovery rate or false non-discovery rate) in large dimensions. Remarkably, degrees of freedom in the chi-square distributions do not affect the boundaries in all four phase transitions. Several well-known procedures are shown to attain these boundaries. Two new phase transitions are also identified in the Gaussian location model under one-sided alternatives. We then elucidate on the nature of signal sizes in association tests by characterizing its relationship with marginal frequencies, odds ratio, and sample sizes in $2\times2$ contingency tables. This allows us to illustrate an interesting manifestation of the phase transition phenomena in genome-wide association studies (GWAS). We also show, perhaps surprisingly, that given total sample sizes, balanced designs in such association studies rarely deliver optimal power for detecting the effects of rare genetic variants.

math.ST↗

arXiv2021

Improving Community Detection Performance in Heterogeneous Music Network by Learning Edge-type Usefulness Distribution

With music becoming an essential part of daily life, there is an urgent need to develop recommendation systems to assist people targeting better songs with fewer efforts. As the interactions between users and songs naturally construct a complex network, community detection approaches can be applied to reveal users' potential interests on songs by grouping relevant users & songs to the same community. However, as the types of interaction could be heterogeneous, it challenges conventional community detection methods designed originally for homogeneous networks. Although there are existing works on heterogeneous community detection, they are mostly task-driven approaches and not feasible for specific music recommendation. In this paper, we propose a genetic based approach to learn an edge-type usefulness distribution (ETUD) for all edge-types in heterogeneous music networks. ETUD can be regarded as a linear function to project all edges to the same latent space and make them comparable. Therefore a heterogeneous network can be converted to a homogeneous one where those conventional methods are eligible to use. We validate the proposed model on a heterogeneous music network constructed from an online music streaming service. Results show that for conventional methods, ETUD can help to detect communities significantly improving music recommendation accuracy while simultaneously reducing user searching cost.

cs.IR↗

arXiv2021

Two-stage Voice Application Recommender System for Unhandled Utterances in Intelligent Personal Assistant

Intelligent personal assistants (IPA) enable voice applications that facilitate people's daily tasks. However, due to the complexity and ambiguity of voice requests, some requests may not be handled properly by the standard natural language understanding (NLU) component. In such cases, a simple reply like "Sorry, I don't know" hurts the user's experience and limits the functionality of IPA. In this paper, we propose a two-stage shortlister-reranker recommender system to match third-party voice applications (skills) to unhandled utterances. In this approach, a skill shortlister is proposed to retrieve candidate skills from the skill catalog by calculating both lexical and semantic similarity between skills and user requests. We also illustrate how to build a new system by using observed data collected from a baseline rule-based system, and how the exposure biases can generate discrepancy between offline and human metrics. Lastly, we present two relabeling methods that can handle the incomplete ground truth, and mitigate exposure bias. We demonstrate the effectiveness of our proposed system through extensive offline experiments. Furthermore, we present online A/B testing results that show a significant boost on user experience satisfaction.

cs.LG↗

arXiv2020

Multi-frequency Point Supported LLRF Front-end for CiADS Wide-bandwidth Application

The China initiative Accelerator Driven System, CiADS, physics design adopts $162.5 \,\mathrm{MHz}$, $325 \,\mathrm{MHz}$, and $650 \,\mathrm{MHz}$ cavities, which are driven by the corresponding radio frequency (RF) power system, requiring frequency translation front-end for the RF station. For that application, a general-purpose design front-end prototype has been developed to evaluate the multi-frequency point supported design feasibility. The difficult parts to achieve the requirements of the general-purpose design are reasonable device selection and balanced design. With a carefully selected low-noise wide-band RF mixer and amplifier to balance the performance of multi-frequency supported down-conversion, specially designed local oscillator (LO) distribution net to increase isolation between adjacent channels, and external band-pass filter to realize expected up-conversion frequencies, high maintenance and modular front-end general-purpose design has been implemented. Results of standard parameters show an $R^2$ value of at least $99.991\%$ in the range of $-60 \sim 10\,\mathrm{dBm}$ for linearity, up to $18\,\mathrm{dBm}$ for P1dB, and up to $89\,\mathrm{dBc}$ for crosstalk between adjacent channels. The phase noise spectrum is lower than $80\,\mathrm{dBc}$ in the range of $0 \sim 1\,\mathrm{MHz}$, and cumulative phase noise is $0.006^\circ$; amplitude and phase stability are $0.022\%$ and $0.034^\circ$, respectively.

physics.acc-ph↗

arXiv2020

Efficient Personalized Community Detection via Genetic Evolution

Personalized community detection aims to generate communities associated with user need on graphs, which benefits many downstream tasks such as node recommendation and link prediction for users, etc. It is of great importance but lack of enough attention in previous studies which are on topics of user-independent, semi-supervised, or top-K user-centric community detection. Meanwhile, most of their models are time consuming due to the complex graph structure. Different from these topics, personalized community detection requires to provide higher-resolution partition on nodes that are more relevant to user need while coarser manner partition on the remaining less relevant nodes. In this paper, to solve this task in an efficient way, we propose a genetic model including an offline and an online step. In the offline step, the user-independent community structure is encoded as a binary tree. And subsequently an online genetic pruning step is applied to partition the tree into communities. To accelerate the speed, we also deploy a distributed version of our model to run under parallel environment. Extensive experiments on multiple datasets show that our model outperforms the state-of-arts with significantly reduced running time.

cs.IR↗