arXiv subjects

Yao Liang

Publications and source records attributed to Yao Liang.

At least 19 recordsLinked to original sources

arXiv2026

Policy, Technology, and Economic Efficiency of Infrastructure Energy Investment: A Strategic Analysis for a Low-Carbon Future

This study provides a comprehensive strategic analysis of infrastructure energy investment in the context of the global low-carbon transition. Integrating quantitative panel data analysis across 15 countries (2010-2023), detailed case studies of Germany, the United States, China, and the European Union, and scenario simulations through 2050, we examine how policy, technology, and economic factors interact to determine investment effectiveness. Using panel data from 15 countries over the period 2010-2023, we find that renewable energy investment is systematically associated with higher economic growth and lower carbon emissions after controlling for country and year fixed effects.

econ.GN↗

arXiv2025

DiverValue-Bench: A Benchmark and Fine-Tuning Framework for Aligning Large Language Models with Diverse Human Values

Aligning large language models (LLMs) with diverse human values is essential for safe and effective deployment, yet existing benchmarks often overlook cultural and demographic variation. We introduce DiverValue-Bench, a population-aware benchmark for evaluating multi-dimensional value alignment across 74 countries/regions. It contains 23,763 quality-controlled instances derived from PRISM user feedback and audited through large-scale human validation, with fine-grained value labels, personalized questions, contrastive reference answers, and rich demographic metadata. Using DiverValue-Bench, we evaluate representative LLMs and reveal substantial geographic and demographic disparities that are masked by aggregate performance. We further show that lightweight preference-based fine-tuning with Low-Rank Adaptation (LoRA) and Direct Preference Optimization (DPO) substantially improves in-domain value alignment while yielding consistent out-of-domain gains. These results highlight the need for population-aware alignment evaluation and demonstrate the utility of DiverValue-Bench as a practical foundation for global alignment, personalized value modeling, and equitable AI development.

cs.CL↗

arXiv2025

Super Co-alignment of Human and AI for Sustainable Symbiotic Society

As Artificial Intelligence (AI) advances toward Artificial General Intelligence (AGI) and eventually Artificial Superintelligence (ASI), it may potentially surpass human control, deviate from human values, and even lead to irreversible catastrophic consequences in extreme cases. This looming risk underscores the critical importance of the "superalignment" problem - ensuring that AI systems which are much smarter than humans, remain aligned with human (compatible) intentions and values. While current scalable oversight and weak-to-strong generalization methods demonstrate certain applicability, they exhibit fundamental flaws in addressing the superalignment paradigm - notably, the unidirectional imposition of human values cannot accommodate superintelligence's autonomy or ensure AGI/ASI's stable learning. We contend that the values for sustainable symbiotic society should be co-shaped by humans and living AI together, achieving "Super Co-alignment." Guided by this vision, we propose a concrete framework that integrates external oversight and intrinsic proactive alignment. External oversight superalignment should be grounded in human-centered ultimate decision, supplemented by interpretable automated evaluation and correction, to achieve continuous alignment with humanity's evolving values. Intrinsic proactive superalignment is rooted in a profound understanding of the Self, others, and society, integrating self-awareness, self-reflection, and empathy to spontaneously infer human intentions, distinguishing good from evil and proactively prioritizing human well-being. The integration of externally-driven oversight with intrinsically-driven proactive alignment will co-shape symbiotic values and rules through iterative human-ASI co-alignment, paving the way for achieving safe and beneficial AGI and ASI for good, for human, and for a symbiotic ecology.

cs.AI↗

arXiv2025

AI Governance InternationaL Evaluation Index (AGILE Index) 2024

The rapid advancement of Artificial Intelligence (AI) technology is profoundly transforming human society and concurrently presenting a series of ethical, legal, and social issues. The effective governance of AI has become a crucial global concern. Since 2022, the extensive deployment of generative AI, particularly large language models, marked a new phase in AI governance. Continuous efforts are being made by the international community in actively addressing the novel challenges posed by these AI developments. As consensus on international governance continues to be established and put into action, the practical importance of conducting a global assessment of the state of AI governance is progressively coming to light. In this context, we initiated the development of the AI Governance InternationaL Evaluation Index (AGILE Index). Adhering to the design principle, "the level of governance should match the level of development," the inaugural evaluation of the AGILE Index commences with an exploration of four foundational pillars: the development level of AI, the AI governance environment, the AI governance instruments, and the AI governance effectiveness. It covers 39 indicators across 18 dimensions to comprehensively assess the AI governance level of 14 representative countries globally. The index is utilized to delve into the status of AI governance to date in 14 countries for the first batch of evaluation. The aim is to depict the current state of AI governance in these countries through data scoring, assist them in identifying their governance stage and uncovering governance issues, and ultimately offer insights for the enhancement of their AI governance systems.

cs.CY↗

arXiv2025

TriAdaptLoRA: Brain-Inspired Triangular Adaptive Low-Rank Adaptation for Parameter-Efficient Fine-Tuning

The fine-tuning of Large Language Models (LLMs) is pivotal for achieving optimal performance across diverse downstream tasks. However, while full fine-tuning delivers superior results, it entails significant computational and resource costs. Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, address these challenges by reducing the number of trainable parameters, but they often struggle with rank adjustment efficiency and task-specific adaptability. We propose Triangular Adaptive Low-Rank Adaptation (TriAdaptLoRA), a novel PEFT framework inspired by neuroscience principles, which dynamically optimizes the allocation of trainable parameters. TriAdaptLoRA introduces three key innovations: 1) a triangular split of transformation matrices into lower and upper triangular components to maximize parameter utilization, 2) a parameter importance metric based on normalized Frobenius norms for efficient adaptation, and 3) an adaptive rank-growth strategy governed by dynamic thresholds, allowing flexible parameter allocation across training steps. Experiments conducted on a variety of natural language understanding and generation tasks demonstrate that TriAdaptLoRA consistently outperforms existing PEFT methods. It achieves superior performance, enhanced stability, and reduced computational overhead, particularly under linear threshold-driven rank growth. These results highlight its efficacy as a scalable and resource-efficient solution for fine-tuning LLMs.

cs.CL↗

arXiv2024

Matrix-Transformation Based Low-Rank Adaptation (MTLoRA): A Brain-Inspired Method for Parameter-Efficient Fine-Tuning

Fine-tuning techniques based on Large Pretrained Language Models (LPLMs) have been proven to significantly enhance model performance on a variety of downstream tasks and effectively control the output behaviors of LPLMs. Recent studies have proposed numerous methods for fine-tuning a small number of parameters based on open-source LPLMs, reducing the demand for computational and storage resources. Among these, reparameterization fine-tuning methods represented by LoRA (Low-Rank Adaptation) have gained popularity. We find that although these methods perform well in many aspects, there is still considerable room for improvement in terms of complex task adaptability, performance, stability, and algorithm complexity. In response to this, inspired by the idea that the functions of the brain are shaped by its geometric structure, this paper integrates this idea into LoRA technology and proposes a new matrix transformation-based reparameterization method for efficient fine-tuning, named Matrix-Transformation based Low-Rank Adaptation (MTLoRA). MTLoRA aims to dynamically alter its spatial geometric structure by applying a transformation-matrix T to perform linear transformations, such as rotation, scaling, and translation, on the task-specific parameter matrix, generating new matrix feature patterns (eigenvectors) to mimic the fundamental influence of complex geometric structure feature patterns in the brain on functions, thereby enhancing the model's performance in downstream tasks. In Natural Language Understanding (NLU) tasks, it is evaluated using the GLUE benchmark test, and the results reveal that MTLoRA achieves an overall performance increase of about 1.0% across eight tasks; in Natural Language Generation (NLG) tasks, MTLoRA improves performance by an average of 0.95% and 0.56% in the DART and WebNLG tasks, respectively.

cs.CL↗

arXiv2023

From local to nonlocal high-Q plasmonic metasurfaces

The physics of bound states in the continuum (BICs) allows to design and demonstrate optical resonant structures with large values of the quality factor ($Q$-factor) by employing dielectric structures with low losses. However, BIC is a general wave phenomenon that should be observed in many systems, including the metal-dielectric structures supporting plasmons where the resonances are hindered by losses. Here we develop a comprehensive strategy to achieve high-$Q$ resonances in plasmonic metasurfaces by effectively tailoring the resonant modes from local and nonlocal regimes.

physics.optics↗

arXiv2023

STREAM: Social data and knowledge collective intelligence platform for TRaining Ethical AI Models

This paper presents Social data and knowledge collective intelligence platform for TRaining Ethical AI Models (STREAM) to address the challenge of aligning AI models with human moral values, and to provide ethics datasets and knowledge bases to help promote AI models "follow good advice as naturally as a stream follows its course". By creating a comprehensive and representative platform that accurately mirrors the moral judgments of diverse groups including humans and AIs, we hope to effectively portray cultural and group variations, and capture the dynamic evolution of moral judgments over time, which in turn will facilitate the Establishment, Evaluation, Embedding, Embodiment, Ensemble, and Evolvement (6Es) of the moral capabilities of AI models. Currently, STREAM has already furnished a comprehensive collection of ethical scenarios, and amassed substantial moral judgment data annotated by volunteers and various popular Large Language Models (LLMs), collectively portraying the moral preferences and performances of both humans and AIs across a range of moral contexts. This paper will outline the current structure and construction of STREAM, explore its potential applications, and discuss its future prospects.

cs.AI↗

arXiv2023

Improving Robustness of Deep Convolutional Neural Networks via Multiresolution Learning

The current learning process of deep learning, regardless of any deep neural network (DNN) architecture and/or learning algorithm used, is essentially a single resolution training. We explore multiresolution learning and show that multiresolution learning can significantly improve robustness of DNN models for both 1D signal and 2D signal (image) prediction problems. We demonstrate this improvement in terms of both noise and adversarial robustness as well as with small training dataset size. Our results also suggest that it may not be necessary to trade standard accuracy for robustness with multiresolution learning, which is, interestingly, contrary to the observation obtained from the traditional single resolution learning setting.

cs.LG↗

arXiv2023

A Brain-inspired Memory Transformation based Differentiable Neural Computer for Reasoning-based Question Answering

Reasoning and question answering as a basic cognitive function for humans, is nevertheless a great challenge for current artificial intelligence. Although the Differentiable Neural Computer (DNC) model could solve such problems to a certain extent, the development is still limited by its high algorithm complexity, slow convergence speed, and poor test robustness. Inspired by the learning and memory mechanism of the brain, this paper proposed a Memory Transformation based Differentiable Neural Computer (MT-DNC) model. MT-DNC incorporates working memory and long-term memory into DNC, and realizes the autonomous transformation of acquired experience between working memory and long-term memory, thereby helping to effectively extract acquired knowledge to improve reasoning ability. Experimental results on bAbI question answering task demonstrated that our proposed method achieves superior performance and faster convergence speed compared to other existing DNN and DNC models. Ablation studies also indicated that the memory transformation from working memory to long-term memory plays essential role in improving the robustness and stability of reasoning. This work explores how brain-inspired memory transformation can be integrated and applied to complex intelligent dialogue and reasoning systems.

cs.AI↗

arXiv2022

Brain-inspired Graph Spiking Neural Networks for Commonsense Knowledge Representation and Reasoning

How neural networks in the human brain represent commonsense knowledge, and complete related reasoning tasks is an important research topic in neuroscience, cognitive science, psychology, and artificial intelligence. Although the traditional artificial neural network using fixed-length vectors to represent symbols has gained good performance in some specific tasks, it is still a black box that lacks interpretability, far from how humans perceive the world. Inspired by the grandmother-cell hypothesis in neuroscience, this work investigates how population encoding and spiking timing-dependent plasticity (STDP) mechanisms can be integrated into the learning of spiking neural networks, and how a population of neurons can represent a symbol via guiding the completion of sequential firing between different neuron populations. The neuron populations of different communities together constitute the entire commonsense knowledge graph, forming a giant graph spiking neural network. Moreover, we introduced the Reward-modulated spiking timing-dependent plasticity (R-STDP) mechanism to simulate the biological reinforcement learning process and completed the related reasoning tasks accordingly, achieving comparable accuracy and faster convergence speed than the graph convolutional artificial neural networks. For the fields of neuroscience and cognitive science, the work in this paper provided the foundation of computational modeling for further exploration of the way the human brain represents commonsense knowledge. For the field of artificial intelligence, this paper indicated the exploration direction for realizing a more robust and interpretable neural network by constructing a commonsense knowledge representation and reasoning spiking neural networks with solid biological plausibility.

cs.NE↗

arXiv2021

Theoretical design study of FWM in silicon nitride waveguides with integrated graphene oxide films

We theoretically investigate and optimize four-wave mixing (FWM) in silicon nitride (SiN) waveguides integrated with two-dimensional (2D) layered graphene oxide (GO) films. Based on extensive previous measurements of the material parameters of the GO films, we perform detailed analysis for the influence of device parameters including waveguide geometry, GO film thickness, length, and coating position on the FWM conversion efficiency (CE) and conversion bandwidth (CB). The influence of dispersion and photo-thermal changes in the GO films is also discussed. Owing to the strong mode overlap between the SiN waveguides and the highly nonlinear GO films, FWM in the hybrid waveguides can be significantly enhanced. We obtain good agreement with previous experimental results and show that by optimizing the device parameters to balance the trade-off between Kerr nonlinearity and loss, the FWM CE can be improved by as much as ~20.7 dB and the FWM CB can be increased by ~4.4 folds, relative to the uncoated waveguides. These results highlight the significantly enhanced FWM performance that can be achieved in SiN waveguides by integrating 2D layered GO films.

physics.optics↗

arXiv2020

Selection of multiple donor gauges via Graphical Lasso for estimation of daily streamflow time series

A fundamental challenge in estimations of daily streamflow time series at sites with incomplete records is how to effectively and efficiently select reference or donor gauges from an existing gauge network to infer the missing data. While research on estimating missing streamflow time series is not new, the existing approaches either use a single reference streamflow gauge or employ a set of "ad-hoc" reference gauges, leaving a systematic selection of reference gauges as a long-standing open question. In this work, a novel method is introduced that facilitates systematical selection of multiple reference gauges from any given streamflow network. The idea is to mathematically characterize the network-wise correlation structure of a streamflow network via graphical Markov modeling, and further transforms a dense network into a sparsely connected one. The resulted underlying sparse graph from the graphical model encodes conditional independence conditions among all reference gauges from the streamflow network, allowing determination of an optimum subset of the donor gauges. The sparsity is discovered by using the Graphical Lasso algorithm with an L1-norm regularization parameter and a thresholding parameter. These two parameters are determined by a multi-objective optimization process. Furthermore, the graphical modeling approach is employed to solve another open problem in gauge removal planning decision (e.g., due to operation budget constraints): which gauges to remove would statistically guarantee the least loss of information by estimations from the remaining gauges? Our graphical model-based method is demonstrated with daily streamflow data from a network of 34 gauges over the Ohio River basin.

stat.AP↗

arXiv2020

High performance integrated polarizers achieved by incorporating 2D layered graphene oxide films

Polarizers and polarization selective resonant cavities (e.g., ring resonators, gratings), are key components for applications to photography, coherent optical detection, polarization-division-multiplexing, optical sensing and liquid crystal displays. We demonstrate waveguide polarizers and polarization discriminating micro-ring resonators (MRRs) by integrating them with 2D graphene oxide (GO) layered thin films. We achieve precise control of the thickness, placement, and size of the films integrated onto photonic devices with a solution based, layer-by-layer transfer-free coating method combined with photolithography and lift-off. This overcomes limitations of layer transfer methods for 2D materials and is a significant advance to manufacturing integrated photonic devices incorporated with 2D materials. We measure the waveguide polarizer for different film thicknesses and lengths versus wavelength, polarization, and power, measuring a high polarization dependent loss (PDL) of ~ 53.8 dB. For GO-coated MRRs, we achieve an extinction ratio difference for TE/TM polarizations of 8.3-dB. We also present measurements of the linear optical properties of 2D layered GO films that yield the material loss anisotropy of the GO films and relative contribution of film loss anisotropy versus polarization-dependent mode overlap. Our results offer interesting physical insights into the transition of the layered GO films from 2D behaviour to quasi bulk like behavior and confirm the high performance of GO based integrated polarization selective devices.

physics.optics↗

arXiv2020

Enhanced four-wave-mixing and 3rd order optical nonlinearity in SiN nanowires integrated with graphene oxide films

Layered 2D graphene oxide (GO) films are integrated with silicon nitride (SiN) waveguides to experimentally demonstrate an enhanced Kerr nonlinearity via four wave mixing (FWM). Owing to the strong light matter interaction between the SiN waveguides and the highly nonlinear GO films, the FWM performance of the hybrid waveguides is significantly improved. SiN waveguides with both uniformly coated and patterned GO films are fabricated based on a transfer free, layer by layer GO coating method together with standard photolithography and lift off processes, yielding precise control of the film thickness, placement and coating length. Detailed FWM measurements are carried out for the fabricated devices with different numbers of GO layers and at different pump powers. By optimizing the trade off between the nonlinearity and loss, we obtain a significant improvement in the FWM conversion efficiency of 7.3 dB for a uniformly coated device with 1 layer of GO and 9.1 dB for a patterned device with 5 layers of GO. We also obtain a significant increase in FWM bandwidth for the patterned devices. A detailed analysis of the influence of pattern length and position on the FWM performance is performed. Based on the FWM measurements, the dependence of GOs third-order nonlinearity on layer number and pump power is also extracted, revealing interesting physical insights about the 2D layered GO films. Finally, we obtain an enhancement in the effective nonlinear parameter of the hybrid waveguides by over a factor of 100. These results verify the enhanced nonlinear optical performance of SiN waveguides achievable by incorporating 2D layered GO films.

physics.optics↗

arXiv2020

Estimation of daily streamflow from multiple donor catchments with Graphical Lasso

A novel algorithm is introduced to improve estimations of daily streamflow time series at sites with incomplete records based on the concept of conditional independence in graphical models. The goal is to fill in gaps of historical data or extend records at streamflow stations no longer in operation or even estimate streamflow at ungauged locations. This is achieved by first selecting relevant stations in the hydrometric network as reference (donor) stations and then using them to infer the missing data. The selection process transforms fully connected streamflow stations in the hydrometric network into a sparsely connected network represented by a precision matrix using a Gaussian graphical model. The underlying graph encodes conditional independence conditions which allow determination of an optimum set of reference stations from the fully connected hydrometric network for a study area. The sparsity of the precision matrix is imposed by using the Graphical Lasso algorithm with an L1-norm regularization parameter and a thresholding parameter. The two parameters are determined by a multi-objective optimization process. In addition, an algorithm based on the conditional independence concept is presented to allow a removal of gauges with the least loss of information. Our approaches are illustrated with daily streamflow data from a hydrometric network of 34 gauges between 1 January 1950 and 31 December 1980 over the Ohio River basin. Our results show that the use of conditional independence conditions can lead to more accurate streamflow estimates than the widely used approaches which are based on either distance or pair-wise correlation.

stat.AP↗

arXiv2019

Integrated polarizers based on graphene oxide in waveguides and ring resonators

Integrated waveguide polarizers and polarization-selective micro-ring resonators (MRRs) incorporated with graphene oxide (GO) films are experimentally demonstrated. CMOS-compatible doped silica waveguides and MRRs with both uniformly coated and patterned GO films are fabricated based on a large-area, transfer-free, layer-by-layer GO coating method that yields precise control of the film thickness. Photolithography and lift-off processes are used to achieve photolithographic patterning of GO films with precise control of the placement and coating length. Detailed measurements are performed to characterize the performance of the devices versus GO film thickness and coating length as a function of polarization, wavelength and power. A high polarization dependent loss of ~53.8 dB is achieved for the waveguide coated with 2-mm-long patterned GO films. It is found that intrinsic film material loss anisotropy dominates the performance for less than 20 layers whereas polarization dependent mode overlap dominates for thicker layers. For the MRRs, the GO coating length is reduced to 50 microns, yielding a ~ 8.3-dB polarization extinction ratio between TE and TM resonances. These results offer interesting physical insights and trends of the layered GO films and demonstrate the effectiveness of introducing GO films into photonic integrated devices to realize high-performance polarization selective components.

physics.optics↗

arXiv2018

Graphene oxide for enhanced nonlinear optics in low and high index contrast waveguides and nanowires

We demonstrate enhanced four-wave mixing (FWM) in doped silica waveguides integrated with graphene oxide (GO) layers. Owing to strong mode overlap between the integrated waveguides and GO films that have a high Kerr nonlinearity and low loss, the FWM efficiency of the hybrid integrated waveguides is significantly improved. We perform FWM measurements for different pump powers, wavelength detuning, GO coating lengths, and number of GO layers. Our experimental results show good agreement with theory, achieving up to ~9.5-dB enhancement in the FWM conversion efficiency for a 1.5-cm-long waveguide integrated with 2 layers of GO. We show theoretically that for different waveguide geometries an enhancement in FWM efficiency of ~ 20 dB can be obtained in the doped silica waveguides, and more than 30 dB in silicon nanowires and slot waveguides. This demonstrates the effectiveness of introducing GO films into integrated photonic devices in order to enhance the performance of nonlinear optical processes.

physics.optics↗