Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization by Majumder, Navonil, Hung, Chia-Yu, Ghosal, Deepanway, Hsu, Wei-Ning, Mihalcea, Rada, Poria, Soujanya

“…The loser outputs, in theory, have some concepts from the prompt missing or in an incorrect order. We fine-tune the publicly available Tango text-to-audio model using diffusion-DPO (direct preference optimization) loss on our preference dataset and show that it leads to improved audio output over Tango and AudioLDM2, in terms of both automatic- and manual-evaluation metrics.…”
Get full text

Defects in β-cell Ca2+ dynamics in age-induced diabetes by Li, Luosheng, Trifunovic, Aleksandra, Köhler, Martin, Wang, Yixin, Petrovic Berglund, Jelena, Illies, Christopher, Juntti-Berggren, Lisa, Larsson, Nils-Göran, Berggren, Per-Olof

“…Our data suggest that aging is associated with a progressive decline in β-cell mitochondrial function that negatively impacts on the fine tuning of Ca2+ dynamics. This is conceptually important since it emphasizes that even relatively modest changes in β-cell signal transduction over time lead to compromised insulin release and a diabetic phenotype.…”
Get full text
Get full text

A two-stage outlier filtering framework for city-scale localization using 3D SfM point clouds by Cheng, Wentao, Chen, Kan, Lin, Weisi, Goesele, Michael, Zhang, Xinfeng, Zhang, Yabin

“…Second, we apply a geometry-based outlier filter to generate a set of fine-grained matches with a novel data-driven geometrical constraint for efficient inlier evaluation. …”
Get full text

Soil liquefaction assessment using soft computing approaches based on capacity energy concept by Chen, Zhixiong, Li, Hongrui, Goh, Anthony Teck Chee, Wu, Chongzhi, Zhang, Wengang

“…Several liquefaction evaluation procedures and approaches have been developed relating the capacity energy to the initial soil parameters, such as the relative density, initial effective confining pressure, fine contents, and soil textural properties. In this study, based on the capacity energy database by Baziar et al. (2011), analyses have been carried out on a total of 405 previously published tests using soft computing approaches, including Ridge, Lasso & LassoCV, Random Forest, eXtreme Gradient Boost (XGBoost), and Multivariate Adaptive Regression Splines (MARS) approaches, to assess the capacity energy required to trigger liquefaction in sand and silty sands. …”
Get full text

Generating domain-specific paraphrases of questions from FAQ by Ng, Jing Rui

“…Firstly, T5 is used to fine-tune on the paraphrase dataset for the task of paraphrase generation. …”
Get full text

Hedonic and non-hedonic bias toward the future by Greene, Preston, Latham, Andrew J., Miller, Kristie, Norton, James

“…Instead, we develop a more fine-grained approach, according to which three factors—positive/negative valence, first/third-person, and hedonic/non-hedonic—each independently influence, but do not determine, whether an event is treated in a future-biased or a time-neutral way. …”
Get full text

Vehicle re-identification using machine learning by Tang, Lisha

“…PANet and PMNet construct a two-stage attention structure to perform a coarse-to-fine search among identities. Finally, we address this Re-ID issue as a multi-task problem and employ Homoscedastic Uncertainty Learning to automatically balance the loss weightings. …”
Get full text

Aspect-based sentiment analysis using BERT by Kheriwala, Hussain Khozema

“…Aspect-based sentiment analysis extracts and identifies fine-grained sentiment polarities for a specific aspect. …”
Get full text

Tackling background ambiguities in multi-class few-shot point cloud semantic segmentation by Lai, Lvlong, Chen, Jian, Zhang, Chi, Zhang, Zehong, Lin, Guosheng, Wu, Qingyao

“…With our proposed approach, fine-tuning, a weak baseline method for few-shot segmentation, gains significant performance improvement (e.g., 7.48% and 7.04% in 2-way-1-shot and 3-way-1-shot tasks of S3DIS, respectively) and outperforms current state-of-the-art methods in all the task settings of S3DIS and ScanNet benchmark datasets.…”
Get full text

PromptChart: prompting InstructGPT for zero & few-shot chart question answering and summarization by Do, Xuan Long

“…Existing state-of-the-art models rely on fine-tuning strategies and struggle with tasks that involve logical and arithmetic reasoning. …”
Get full text

Effect of heat treatment on microstructures and mechanical properties of Inconel 718 additively manufactured using gradient laser power by Xu, Luming, Chai, Ze, Peng, Bo, Zhou, Wei, Chen, Xiaoqi

“…The as-deposited GLP samples with fine discrete Laves phases and a low SR can achieve a relatively uniform element distribution and γ′′ and γ′ phase precipitation after standard solution treatment plus aging (STA), while CLP samples with coarse long-chain Laves phases and high SR require a high-temperature homogenization plus STA (HSTA) heat treatment to achieve the uniform element distribution. …”
Get full text

Unsupervised land-use change detection using multi-temporal POI embedding by Yao, Yao, Zhu, Qia, Guo, Zijin, Huang, Weiming, Zhang, Yatao, Yan, Xiaoqin, Dong, Anning, Jiang, Zhangwei, Liu, Hong, Guan, Qingfeng

“…In addition, our model can effectively identify areas with unchanged land use and land use changes in residential and industrial areas at a fine scale.…”
Get full text

Compositionally tuned hybridization of n-type Ag0: Ag2Se under ambient conditions towards excellent thermoelectric properties at room temperature by Tee, Si Yin, Ponsford, Daniel, Tan, Xian Yi, Wang, Xiaobai, Lay, Chee Leng, Lee, Coryl Jing Jun, Ni, Xi Ping, Seng, Debbie Hwee Leng, Thitsartarn, Warintorn, Guan, Guijian, Han, Ming-Yong

“…This strategy enables fine control over phases and compositions through nanoscale hybridization using silver nanoparticles for the first time, yet remains applicable to large-scale production methods. …”
Get full text

Suppressing condensation frosting using micropatterned ice walls by Zuo, Zichao, Zhao, Yugang, Li, Kang, Zhang, Hua, Yang, Chun

“…To ensure vapor preferentially depositing on upper sidewalls, a critical ice wall height exists, below which fine ice grains initiate in the adjacent of the substrate base and grow substantially faster, compromising the anti-frosting performance. …”
Get full text

An enhanced deep reinforcement learning ensemble empowered by large language model by Li, Xinyi

“…At its core, LLM-DRE effectively processes and synthesizes disparate forms of market data to fine-tune trading strategies, enabling an enriched assessment of financial environments. …”
Get full text

Annotating videos that teach MS Excel and predicting mouse / keyboard actions by Tan, Genson Yao Jie

“…The research methodology involves exploring the limitations of fine-tuning Flan-T5 for this task, while also utilizing prompt engineering on Large Language Model (LLM) such as Llama 2 as an alternative method. …”
Get full text

Deep learning for fabric defect detection by Nangia, Saniya

“…The base models, YOLOv8 and RT-DETR, are fine-tuned on a dataset of patterned fabric defects to find the optimal model configurations. …”
Get full text

3D-printed mechano-optic force sensor for soft robotic gripper enabled by programmable structural metamaterials by Hegde, Chidanand, Mysa, Ravi Chaithanya, Chooi, Aaron, Dontu, Saikrishna, Tan, Joel Ming Rui, Wong, Lydia Helena, Valdivia y Alvarado, Pablo, Magdassi, Shlomo

“…The design supports seamless integration of the sensor into soft grippers, enabling 3D printing of the soft gripper with an embedded sensor in a single step, thus overcoming the tedious and complex and multiple fabrication steps commonly applied in conventional processes. The sensor boasts a fine resolution of 0.015 N, a measurement range up to 16 N, linearity (adj. …”
Get full text

ContrastSense: domain-invariant contrastive learning for in-the-wild wearable sensing by Dai, Gaole, Xu, Huatao, Yoon, Hyungjun, Li, Mo, Tan, Rui, Lee, Sung-Ju

“…In addition, ContrastSense designs a parameter-wise penalty to preserve domaininvariant knowledge during fine-tuning to further maintain model robustness. …”
Get full text

Hybrid superplastic forming of magnesium alloys by Sze, Swai Ming

“…However, this technology requires expensive materials with fine grain sizes and high temperature conditions with slow forming rates. …”
Get full text