arXiv subjects

Yong Cao

Publications and source records attributed to Yong Cao.

At least 37 records · Page 2Linked to original sources

arXiv2025

Scholar Inbox: Personalized Paper Recommendations for Scientists

Scholar Inbox is a new open-access platform designed to address the challenges researchers face in staying current with the rapidly expanding volume of scientific literature. We provide personalized recommendations, continuous updates from open-access archives (arXiv, bioRxiv, etc.), visual paper summaries, semantic search, and a range of tools to streamline research workflows and promote open research access. The platform's personalized recommendation system is trained on user ratings, ensuring that recommendations are tailored to individual researchers' interests. To further enhance the user experience, Scholar Inbox also offers a map of science that provides an overview of research across domains, enabling users to easily explore specific topics. We use this map to address the cold start problem common in recommender systems, as well as an active learning strategy that iteratively prompts users to rate a selection of papers, allowing the system to learn user preferences quickly. We evaluate the quality of our recommendation system on a novel dataset of 800k user ratings, which we make publicly available, as well as via an extensive user study. https://www.scholar-inbox.com/

cs.CL↗

arXiv2025

Full realization of the RIBLL2 separator at the HIRFL-CSR facility

A new experimental platform was constructed at the Second Radioactive Ion Beam Line in Lanzhou (RIBLL2) of HIRFL-CSR accelerator facility at Lanzhou, China. Its performance, along with several newly developed detectors, was tested in two radioactive ion beam experiments utilizing a 400 MeV/u 40Ar beam and a 350 MeV/u 78Kr beam, respectively. The first results from these two experiments demonstrate a good particle identification capability of the setup, thereby affirming the full realization of the RIBLL2 separator.

physics.ins-det↗

arXiv2025

QuaDMix: Quality-Diversity Balanced Data Selection for Efficient LLM Pretraining

Quality and diversity are two critical metrics for the training data of large language models (LLMs), positively impacting performance. Existing studies often optimize these metrics separately, typically by first applying quality filtering and then adjusting data proportions. However, these approaches overlook the inherent trade-off between quality and diversity, necessitating their joint consideration. Given a fixed training quota, it is essential to evaluate both the quality of each data point and its complementary effect on the overall dataset. In this paper, we introduce a unified data selection framework called QuaDMix, which automatically optimizes the data distribution for LLM pretraining while balancing both quality and diversity. Specifically, we first propose multiple criteria to measure data quality and employ domain classification to distinguish data points, thereby measuring overall diversity. QuaDMix then employs a unified parameterized data sampling function that determines the sampling probability of each data point based on these quality and diversity related labels. To accelerate the search for the optimal parameters involved in the QuaDMix framework, we conduct simulated experiments on smaller models and use LightGBM for parameters searching, inspired by the RegMix method. Our experiments across diverse models and datasets demonstrate that QuaDMix achieves an average performance improvement of 7.2% across multiple benchmarks. These results outperform the independent strategies for quality and diversity, highlighting the necessity and ability to balance data quality and diversity.

cs.CL↗

arXiv2025

Specializing Large Language Models to Simulate Survey Response Distributions for Global Populations

Large-scale surveys are essential tools for informing social science research and policy, but running surveys is costly and time-intensive. If we could accurately simulate group-level survey results, this would therefore be very valuable to social science research. Prior work has explored the use of large language models (LLMs) for simulating human behaviors, mostly through prompting. In this paper, we are the first to specialize LLMs for the task of simulating survey response distributions. As a testbed, we use country-level results from two global cultural surveys. We devise a fine-tuning method based on first-token probabilities to minimize divergence between predicted and actual response distributions for a given question. Then, we show that this method substantially outperforms other methods and zero-shot classifiers, even on unseen questions, countries, and a completely unseen survey. While even our best models struggle with the task, especially on unseen questions, our results demonstrate the benefits of specialization for simulation, which may accelerate progress towards sufficiently accurate simulation in the future.

cs.CL↗

arXiv2025

Does Mapo Tofu Contain Coffee? Probing LLMs for Food-related Cultural Knowledge

Recent studies have highlighted the presence of cultural biases in Large Language Models (LLMs), yet often lack a robust methodology to dissect these phenomena comprehensively. Our work aims to bridge this gap by delving into the Food domain, a universally relevant yet culturally diverse aspect of human life. We introduce FmLAMA, a multilingual dataset centered on food-related cultural facts and variations in food practices. We analyze LLMs across various architectures and configurations, evaluating their performance in both monolingual and multilingual settings. By leveraging templates in six different languages, we investigate how LLMs interact with language-specific and cultural knowledge. Our findings reveal that (1) LLMs demonstrate a pronounced bias towards food knowledge prevalent in the United States; (2) Incorporating relevant cultural context significantly improves LLMs' ability to access cultural knowledge; (3) The efficacy of LLMs in capturing cultural nuances is highly dependent on the interplay between the probing language, the specific model architecture, and the cultural context in question. This research underscores the complexity of integrating cultural understanding into LLMs and emphasizes the importance of culturally diverse datasets to mitigate biases and enhance model performance across different cultural domains.

cs.CL↗

arXiv2024

Vote&Mix: Plug-and-Play Token Reduction for Efficient Vision Transformer

Despite the remarkable success of Vision Transformers (ViTs) in various visual tasks, they are often hindered by substantial computational cost. In this work, we introduce Vote\&Mix (\textbf{VoMix}), a plug-and-play and parameter-free token reduction method, which can be readily applied to off-the-shelf ViT models \textit{without any training}. VoMix tackles the computational redundancy of ViTs by identifying tokens with high homogeneity through a layer-wise token similarity voting mechanism. Subsequently, the selected tokens are mixed into the retained set, thereby preserving visual information. Experiments demonstrate VoMix significantly improves the speed-accuracy tradeoff of ViTs on both images and videos. Without any training, VoMix achieves a 2$\times$ increase in throughput of existing ViT-H on ImageNet-1K and a 2.4$\times$ increase in throughput of existing ViT-L on Kinetics-400 video dataset, with a mere 0.3\% drop in top-1 accuracy.

cs.CV↗

arXiv2024

Utility of High-Order Scheme for Unsteady Flow Simulations: Comparison with Second-Order Tool

The objective of this work is to investigate the utility and effectiveness of the high-order scheme for simulating unsteady turbulent flows. To achieve it, the studies were conducted from two perspectives: (i) the ability of different numerical schemes for turbulence problems under the same set of meshes; and (ii) the accuracy and stability of higher-order schemes for solving turbulence statistics for different mesh types (hexahedral, tetrahedral, and polyhedral cells). The simulations employ the third-order scheme for spatial discretization of the governing equations, while a widely-used second-order solver, namely pisoFoam, was employed for comparison. This study considers the canonical cases of the Taylor-Green vortex (TGV) problem at Re=100, 1600 and flow past a sphere at Re=10000 to address the aforementioned two key issues. For the TGV case, the high-order model significantly improves the numerical accuracy with convergence rates and reduces the numerical dissipation of nearly 1/10 of pisoFoam. In the latter case, the high-order scheme with large-eddy simulation (LES) accurately predicts the vortex structures and the flow instability, regardless of grid type. However, pisoFoam is found to be sensitive to mesh types, which results in numerous non-physical structures in the flow field due to numerical noise rather than flow physics, particularly for tetrahedral cells. Furthermore, for the typical low- and high-order flow statistics, the numerical results predicted by the present model show better agreement with the reference data and have less dependence on the type of grids compared with the conventional scheme. In addition, the obtained energy spectrum by the high-order solver accurately captures the Kelvin-Helmholtz (K-H) instability and the vortex shedding frequency, while these important features are less pronounced by the traditional low-order model.

physics.flu-dyn↗

arXiv2024

High-flexibility reconstruction of small-scale motions in wall turbulence using a generalized zero-shot learning

This study proposes a novel super-resolution (or SR) framework for generating high-resolution turbulent boundary layer (TBL) flow from low-resolution inputs. The framework combines a super-resolution generative adversarial neural network (SRGAN) with down-sampling modules (DMs), integrating the residual of the continuity equation into the loss function. DMs selectively filter out components with excessive energy dissipation in low-resolution fields prior to the super-resolution process. The framework iteratively applies the SRGAN and DM procedure to fully capture the energy cascade of multi-scale flow structures, collectively termed the SRGAN-based energy cascade framework (EC-SRGAN). Despite being trained solely on turbulent channel flow data (via "zero-shot transfer"), EC-SRGAN exhibits remarkable generalization in predicting TBL small-scale velocity fields, accurately reproducing wavenumber spectra compared to DNS results. Furthermore, a super-resolution core is trained at a specific super-resolution ratio. By leveraging this pre-trained super-resolution core, EC-SRGAN efficiently reconstructs TBL fields at multiple super-resolution ratios from various levels of low-resolution inputs, showcasing strong flexibility. By learning turbulent scale invariance, EC-SRGAN demonstrates robustness across different TBL datasets. These results underscore EC-SRGAN potential for generating and predicting wall turbulence with high flexibility, offering promising applications in addressing diverse TBL-related challenges.

physics.flu-dyn↗

arXiv2024

Vision-Language Models under Cultural and Inclusive Considerations

Large vision-language models (VLMs) can assist visually impaired people by describing images from their daily lives. Current evaluation datasets may not reflect diverse cultural user backgrounds or the situational context of this use case. To address this problem, we create a survey to determine caption preferences and propose a culture-centric evaluation benchmark by filtering VizWiz, an existing dataset with images taken by people who are blind. We then evaluate several VLMs, investigating their reliability as visual assistants in a culturally diverse setting. While our results for state-of-the-art models are promising, we identify challenges such as hallucination and misalignment of automatic evaluation metrics with human judgment. We make our survey, data, code, and model outputs publicly available.

cs.CV↗

arXiv2024

Rethinking the Effectiveness of Graph Classification Datasets in Benchmarks for Assessing GNNs

Graph classification benchmarks, vital for assessing and developing graph neural networks (GNNs), have recently been scrutinized, as simple methods like MLPs have demonstrated comparable performance. This leads to an important question: Do these benchmarks effectively distinguish the advancements of GNNs over other methodologies? If so, how do we quantitatively measure this effectiveness? In response, we first propose an empirical protocol based on a fair benchmarking framework to investigate the performance discrepancy between simple methods and GNNs. We further propose a novel metric to quantify the dataset effectiveness by considering both dataset complexity and model performance. To the best of our knowledge, our work is the first to thoroughly study and provide an explicit definition for dataset effectiveness in the graph learning area. Through testing across 16 real-world datasets, we found our metric to align with existing studies and intuitive assumptions. Finally, we explore the causes behind the low effectiveness of certain datasets by investigating the correlation between intrinsic graph properties and class labels, and we developed a novel technique supporting the correlation-controllable synthetic dataset generation. Our findings shed light on the current understanding of benchmark datasets, and our new platform could fuel the future evolution of graph classification benchmarks.

cs.LG↗

arXiv2024

Exploring Visual Culture Awareness in GPT-4V: A Comprehensive Probing

Pretrained large Vision-Language models have drawn considerable interest in recent years due to their remarkable performance. Despite considerable efforts to assess these models from diverse perspectives, the extent of visual cultural awareness in the state-of-the-art GPT-4V model remains unexplored. To tackle this gap, we extensively probed GPT-4V using the MaRVL benchmark dataset, aiming to investigate its capabilities and limitations in visual understanding with a focus on cultural aspects. Specifically, we introduced three visual related tasks, i.e. caption classification, pairwise captioning, and culture tag selection, to systematically delve into fine-grained visual cultural evaluation. Experimental results indicate that GPT-4V excels at identifying cultural concepts but still exhibits weaker performance in low-resource languages, such as Tamil and Swahili. Notably, through human evaluation, GPT-4V proves to be more culturally relevant in image captioning tasks than the original MaRVL human annotations, suggesting a promising solution for future visual cultural benchmark construction.

cs.CL↗

arXiv2024

Geo-Encoder: A Chunk-Argument Bi-Encoder Framework for Chinese Geographic Re-Ranking

Chinese geographic re-ranking task aims to find the most relevant addresses among retrieved candidates, which is crucial for location-related services such as navigation maps. Unlike the general sentences, geographic contexts are closely intertwined with geographical concepts, from general spans (e.g., province) to specific spans (e.g., road). Given this feature, we propose an innovative framework, namely Geo-Encoder, to more effectively integrate Chinese geographical semantics into re-ranking pipelines. Our methodology begins by employing off-the-shelf tools to associate text with geographical spans, treating them as chunking units. Then, we present a multi-task learning module to simultaneously acquire an effective attention matrix that determines chunk contributions to extra semantic representations. Furthermore, we put forth an asynchronous update mechanism for the proposed addition task, aiming to guide the model capable of effectively focusing on specific chunks. Experiments on two distinct Chinese geographic re-ranking datasets, show that the Geo-Encoder achieves significant improvements when compared to state-of-the-art baselines. Notably, it leads to a substantial improvement in the Hit@1 score of MGEO-BERT, increasing it by 6.22% from 62.76 to 68.98 on the GeoTES dataset.

cs.CL↗

arXiv2024

Bridging Cultural Nuances in Dialogue Agents through Cultural Value Surveys

The cultural landscape of interactions with dialogue agents is a compelling yet relatively unexplored territory. It's clear that various sociocultural aspects -- from communication styles and beliefs to shared metaphors and knowledge -- profoundly impact these interactions. To delve deeper into this dynamic, we introduce cuDialog, a first-of-its-kind benchmark for dialogue generation with a cultural lens. We also develop baseline models capable of extracting cultural attributes from dialogue exchanges, with the goal of enhancing the predictive accuracy and quality of dialogue agents. To effectively co-learn cultural understanding and multi-turn dialogue predictions, we propose to incorporate cultural dimensions with dialogue encoding features. Our experimental findings highlight that incorporating cultural value surveys boosts alignment with references and cultural markers, demonstrating its considerable influence on personalization and dialogue quality. To facilitate further exploration in this exciting domain, we publish our benchmark publicly accessible at https://github.com/yongcaoplus/cuDialog.

cs.CL↗

arXiv2024

MLPs Compass: What is learned when MLPs are combined with PLMs?

While Transformer-based pre-trained language models and their variants exhibit strong semantic representation capabilities, the question of comprehending the information gain derived from the additional components of PLMs remains an open question in this field. Motivated by recent efforts that prove Multilayer-Perceptrons (MLPs) modules achieving robust structural capture capabilities, even outperforming Graph Neural Networks (GNNs), this paper aims to quantify whether simple MLPs can further enhance the already potent ability of PLMs to capture linguistic information. Specifically, we design a simple yet effective probing framework containing MLPs components based on BERT structure and conduct extensive experiments encompassing 10 probing tasks spanning three distinct linguistic levels. The experimental results demonstrate that MLPs can indeed enhance the comprehension of linguistic structure by PLMs. Our research provides interpretable and valuable insights into crafting variations of PLMs utilizing MLPs for tasks that emphasize diverse linguistic structures.

cs.CL↗

arXiv2023

Cultural Adaptation of Recipes

Building upon the considerable advances in Large Language Models (LLMs), we are now equipped to address more sophisticated tasks demanding a nuanced understanding of cross-cultural contexts. A key example is recipe adaptation, which goes beyond simple translation to include a grasp of ingredients, culinary techniques, and dietary preferences specific to a given culture. We introduce a new task involving the translation and cultural adaptation of recipes between Chinese and English-speaking cuisines. To support this investigation, we present CulturalRecipes, a unique dataset comprised of automatically paired recipes written in Mandarin Chinese and English. This dataset is further enriched with a human-written and curated test set. In this intricate task of cross-cultural recipe adaptation, we evaluate the performance of various methods, including GPT-4 and other LLMs, traditional machine translation, and information retrieval techniques. Our comprehensive analysis includes both automatic and human evaluation metrics. While GPT-4 exhibits impressive abilities in adapting Chinese recipes into English, it still lags behind human expertise when translating English recipes into Chinese. This underscores the multifaceted nature of cultural adaptations. We anticipate that these insights will significantly contribute to future research on culturally-aware language models and their practical application in culturally diverse contexts.

cs.CL↗

arXiv2023

An optimization framework for wind farm layout design using CFD-based Kriging model

Wind farm layout optimization (WFLO) seeks to alleviate the wake loss and maximize wind farm power output efficiency, and is a crucial process in the design of wind energy projects.Since the optimization algorithms typically require thousands of numerical evaluations of the wake effects, conventional WFLO studies are usually carried out with the low-fidelity analytical wake models.In this paper, we develop an optimization framework for wind farm layout design using CFD-based Kriging model to maximize the annual energy production (AEP) of wind farms. This surrogate-based optimization (SBO) framework uses latin hypercube sampling to generate a group of wind farm layout samples, based on which CFD simulations are carried out to obtain the corresponding AEPs.This wind farm layout dataset is used to train the Kriging model, which is then integrated with an optimizer based on genetic algorithm (GA). As the optimization progresses, the intermediate optimal layout designs are again fed into the dataset.Such adaptive update of wind farm layout dataset continues until the algorithm converges.To evaluate the performance of the proposed SBO framework, we apply it to three representative wind farm cases.Compared to the conventional staggered layout, the optimized wind farm produces significantly higher total AEP.In particular, the SBO framework requires significantly smaller number of CFD calls to yield the optimal layouts that generates almost the same AEP with the direct CFD-GA method.Further analysis on the velocity fields show that the optimization framework attempts to locate the downstream turbines away from the the wakes of upstream ones.The proposed CFD-based surrogate model provides a more accurate and flexible alternative to the conventional analytical-wake-model-based methods in WFLO tasks, and has the potential to be used for designing efficient wind farm projects.

physics.flu-dyn↗

arXiv2023

Charge-changing cross section measurements of 300 MeV/nucleon $^{28}$Si on carbon and data analysis

Charge-changing cross section ($σ_{\text{cc}}$) measurements via the transmission method have made important progress recently aiming to determine the charge radii of exotic nuclei. In this work, we report a new $σ_{\text{cc}}$ measurement of 304(9) MeV/nucleon $^{28}$Si on carbon at the second Radioactive Ion Beam Line in Lanzhou (RIBLL2) and describe the data analysis procedure in detail. This procedure is essential to evaluate the systematic uncertainty in the transmission method. The determined $σ_{\mathrm{cc}}$ of 1125(11) mb is found to be consistent with the existing data at similar energies. The present work will serve as a reference in the $σ_{\text{cc}}$ determinations at RIBLL2.

nucl-ex↗

arXiv2023

Explore More Guidance: A Task-aware Instruction Network for Sign Language Translation Enhanced with Data Augmentation

Sign language recognition and translation first uses a recognition module to generate glosses from sign language videos and then employs a translation module to translate glosses into spoken sentences. Most existing works focus on the recognition step, while paying less attention to sign language translation. In this work, we propose a task-aware instruction network, namely TIN-SLT, for sign language translation, by introducing the instruction module and the learning-based feature fuse strategy into a Transformer network. In this way, the pre-trained model's language ability can be well explored and utilized to further boost the translation performance. Moreover, by exploring the representation space of sign language glosses and target spoken language, we propose a multi-level data augmentation scheme to adjust the data distribution of the training set. We conduct extensive experiments on two challenging benchmark datasets, PHOENIX-2014-T and ASLG-PC12, on which our method outperforms former best solutions by 1.65 and 1.42 in terms of BLEU-4. Our code is published at https://github.com/yongcaoplus/TIN-SLT.

cs.CL↗