DR-NTU (Data) (Nanyang Technological University)
DR-NTU (Data) (Nanyang Technological University)Not a member yet
1955 research outputs found
Sort by
Replication Data for: Satellite radar advances carbon emissions accountability over tropical peat
No full textThis data repository contains peat motions derived from ALOS-2 InSAR in Central Kalimantan, Indonesia in the study "Satellite radar advances carbon emissions accountability over tropical peat". Refer to README.txt for further details
Related Data for: Topological optical skyrmion transfer to matter
No full textSource file for Topological optical skyrmion transfer to matte
Skin-inspired flexible and printed iontronic sensor enables bimodal sensing of robot skin for machine-learning-assisted object recognition
No full textCorresponding dat
Replication Data for: Aqueous Colloidal Perovskite Quantum Emitters
No full textAqueous solutions of nanoparticles are the cornerstones for applications in diagnostics, catalysis and more, where control over the nanoparticle's dispersion is pivotal to tailoring the final product properties. Of late, halide perovskite nanocrystals (HPNCs) with outstanding optoelectronic properties emerge as a class of semiconductor nanocrystals distinct from the incumbents. However, HPNCs are particularly susceptible to moisture induced degradation, limiting their utility and regulation in aqueous environments. Here, this hurdle is overcome to realize stable, monodisperse, highly emissive HPNCs in aqueous environments even under ultra-dilute conditions. These colloidal HPNCs are synthesized via a facile room-temperature structural transformation-induced in situ core-shell self-assembly mechanism in contrast to the widely used pre-core-shell approach. The green HPNCs exhibit > 80% photoluminescence quantum yield (PLQY) with excellent water dispersion stability (i.e., zeta potential > 80 mV) even after > 10,000 h in water. Unprecedented aqueous solution phase single‐photon emission with g(2)(0) < 0.2 at concentrations as low as ≈ 0.1 nM is measured. These aqueous HPNCs offer full color tunability that covers the entire Rec. 2020 standard. These findings will lay the foundation for innovative applications of HPNCs in aqueous environments, unlocking new opportunities for nanoscale sensing and optofluidics in photonics, environmental science, and materials engineering
Towards Cross-Modality Modeling for Time Series Analytics: A Survey in the LLM Era
No full textThe proliferation of edge devices has generated an unprecedented volume of time series data across different domains, motivating various well-customized methods. Recently, Large Language Models (LLMs) have emerged as a new paradigm for time series analytics by leveraging the shared sequential nature of textual data and time series. However, a fundamental cross-modality gap between time series and LLMs exists, as LLMs are pre-trained on textual corpora and are not inherently optimized for time series. Many recent proposals are designed to address this issue. In this survey, we provide an up-to-date overview of LLMs-based cross-modality modeling for time series analytics. We first introduce a taxonomy that classifies existing approaches into four groups based on the type of textual data employed for time series modeling. We then summarize key cross-modality strategies, e.g., alignment and fusion, and discuss their applications across a range of downstream tasks. Furthermore, we conduct experiments on multimodal datasets from different application domains to investigate effective combinations of textual data and cross-modality strategies for enhancing time series analytics. Finally, we suggest several promising directions for future research. This survey is designed for a range of professionals, researchers, and practitioners interested in LLM-based time series modeling
F-LMM: Grounding Frozen Large Multimodal Models
No full textEndowing Large Multimodal Models (LMMs) with visual grounding capability can significantly enhance AIs’ understanding of the visual world and their interaction with humans. However, existing methods typically fine-tune the parameters of LMMs to learn additional segmentation tokens and overfit grounding and segmentation datasets. Such a design would inevitably cause a catastrophic diminution in the indispensable conversational capability of general AI assistants. In this paper, we comprehensively evaluate state-of-the-art grounding LMMs across a suite
of multimodal question-answering benchmarks, observing
drastic performance drops that indicate vanishing general knowledge comprehension and weakened instruction
following ability. To address this issue, we present FLMM—grounding frozen off-the-shelf LMMs in human-AI
conversations—a straightforward yet effective design based
on the fact that word-pixel correspondences conducive to
visual grounding inherently exist in the attention mechanism
of well-trained LMMs. Using only a few trainable CNN layers, we can translate word-pixel attention weights to mask
logits, which a SAM-based mask refiner can further optimise. Our F-LMM neither learns special segmentation tokens nor utilises high-quality grounded instruction-tuning
data, but achieves competitive performance on referring
expression segmentation and panoptic narrative grounding benchmarks while completely preserving LMMs’ original conversational ability. Additionally, with instructionfollowing ability preserved and grounding ability obtained,
F-LMM can be directly applied to complex tasks like reasoning segmentation, grounded conversation generation
and visual chain-of-thought reasoning. Our code can be
found at https://github.com/wusize/F-LMM
Data for: Partitioning of Rubisco activase into the pyrenoidal Rubisco condensate is mediated by a functional protein-protein interaction
No full textOriginal data files analyzed for this publication are provided according to the Figure panel in the paper
Computational Data of Wash-Free Bioimaging Review
No full textThe electronic structures and excited-state properties were derived using computational data obtained from density functional theory (DFT) and time-dependent density functional theory (TD-DFT) calculations. All computational studies were conducted using Gaussian 16 software. The results, as discussed in the review paper, were employed to analyze the mechanistic aspects of fluorescence quenching in relevant systems. The computational methods have been highlighted in the corresponding figures, and details are provided in the figure captions to ensure clarity and reproducibility
Replication Data for: Two Causally Related Needles in a Video Haystack
No full textCausal2Needles is a benchmark dataset and evaluation toolkit designed to assess the capabilities of both proprietary and open-source multimodal large language models in long-video understanding. It features a large number of "2-needle" questions, where the model must locate and reason over two distinct pieces of information from the video
Replication Data for: Skyrmionic polarization texture around the phase singularity of optical vortices
No full textExperimental measurement of transverse-axial polarization texture