arXiv subjects

Xiang Dai

Publications and source records attributed to Xiang Dai.

At least 19 recordsLinked to original sources

arXiv2026

Plenoptic imaging of particle interactions in scintillation detectors

Accurate 3D localization of radiation interactions in scintillation detectors is essential for nuclear and particle physics, safeguards, and medical imaging, but remains difficult in light-starved regimes with limited photon statistics. We present PRISM, a multifocal plenoptic imaging system designed for millimeter-scale 3D position reconstruction in a single-volume scintillator. PRISM uses a multifocal microlens array with diverse focal lengths and high effective numerical aperture to balance photon collection with spatial and depth encoding. A Cram'er--Rao lower bound analysis shows that the multifocal design improves axial sensitivity over conventional unifocal plenoptic systems under photon-limited conditions. We build a prototype system, calibrate its optical response with a tunable light source, and form photon-limited measurements with $\mathcal{O}(100)$ detected photons. For sparse single-vertex events, we reconstruct interaction locations using an Alternating Descent Conditional Gradient-inspired algorithm and demonstrate an average 3D localization error of approximately 1 mm. We also provide an initial evaluation of double-vertex events, showing that localization improves as the axial separation between interactions increases. These results demonstrate that multifocal plenoptic imaging can mitigate the traditional trade-off between light collection and spatial resolution, providing a photon-efficient approach to 3D reconstruction in scintillation detectors and a foundation for future multi-scattering event reconstruction.

physics.ins-det↗

arXiv2026

Reliable Thinking with Images

As a multimodal extension of Chain-of-Thought (CoT), Thinking with Images (TWI) has recently emerged as a promising avenue to enhance the reasoning capability of Multi-modal Large Language Models (MLLMs), which generates interleaved CoT by incorporating visual cues into the textual reasoning process. However, the success of existing TWI methods heavily relies on the assumption that interleaved image-text CoTs are faultless, which is easily violated in real-world scenarios due to the complexity of multimodal understanding. In this paper, we reveal and study a highly-practical yet under-explored problem in TWI, termed Noisy Thinking (NT). Specifically, NT refers to the imperfect visual cues mining and answer reasoning process. As the saying goes, ``One mistake leads to another'', erroneous interleaved CoT would cause error accumulation, thus significantly degrading the performance of MLLMs. To solve the NT problem, we propose a novel method dubbed Reliable Thinking with Images (RTWI). In brief, RTWI estimates the reliability of visual cues and textual CoT in a unified text-centric manner and accordingly employs robust filtering and voting modules to prevent NT from contaminating the final answer. Extensive experiments on seven benchmarks verify the effectiveness of RTWI against NT.

cs.CV↗

arXiv2025

Predicting the single-site and multi-site event discrimination power of dual-phase time projection chambers

Dual-phase xenon time projection chambers (TPCs) are widely used in searches for rare dark matter and neutrino interactions, in part because of their excellent position reconstruction capability in 3D. Despite their millimeter-scale resolution along the charge drift axis, xenon TPCs face challenges in resolving single-site (SS) and multi-site (MS) interactions in the transverse plane. In this paper, we build a generic TPC model with an idealized signal readout, and use Fisher Information (FI) to predict its theoretical capability of differentiating SS and MS events using the electroluminescence signal. We also demonstrate via simulation that, when only statistical photon noise is present, the theoretical limits can be approached with conventional reconstruction algorithms like maximum likelihood estimation, and with a convolutional neural network classifier. The implications of this study on future TPC experiments will be discussed.

hep-ex↗

arXiv2025

RnGCam: High-speed video from rolling & global shutter measurements

Compressive video capture encodes a short high-speed video into a single measurement using a low-speed sensor, then computationally reconstructs the original video. Prior implementations rely on expensive hardware and are restricted to imaging sparse scenes with empty backgrounds. We propose RnGCam, a system that fuses measurements from low-speed consumer-grade rolling-shutter (RS) and global-shutter (GS) sensors into video at kHz frame rates. The RS sensor is combined with a pseudorandom optic, called a diffuser, which spatially multiplexes scene information. The GS sensor is coupled with a conventional lens. The RS-diffuser provides low spatial detail and high temporal detail, complementing the GS-lens system's high spatial detail and low temporal detail. We propose a reconstruction method using implicit neural representations (INR) to fuse the measurements into a high-speed video. Our INR method separately models the static and dynamic scene components, while explicitly regularizing dynamics. In simulation, we show that our approach significantly outperforms previous RS compressive video methods, as well as state-of-the-art frame interpolators. We validate our approach in a dual-camera hardware setup, which generates 230 frames of video at 4,800 frames per second for dense scenes, using hardware that costs $10\times$ less than previous compressive video systems.

eess.IV↗

arXiv2025

Single-shot HDR using conventional image sensor shutter functions and optical randomization

High-dynamic-range (HDR) imaging is an essential technique for overcoming the dynamic range limits of image sensors. The classic method relies on multiple exposures, which slows capture time, resulting in motion artifacts when imaging dynamic scenes. Single-shot HDR imaging alleviates this issue by encoding HDR data into a single exposure, then computationally recovering it. Many established methods use strong image priors to recover improperly exposed image detail. These approaches struggle with extended highlight regions. We utilize the global reset release (GRR) shutter mode of an off-the-shelf sensor. GRR shutter mode applies a longer exposure time to rows closer to the bottom of the sensor. We use optics that relay a randomly permuted (shuffled) image onto the sensor, effectively creating spatially randomized exposures across the scene. The exposure diversity allows us to recover HDR data by solving an optimization problem with a simple total variation image prior. In simulation, we demonstrate that our method outperforms other single-shot methods when many sensor pixels are saturated (10% or more), and is competitive at a modest saturation (1%). Finally, we demonstrate a physical lab prototype that uses an off-the-shelf random fiber bundle for the optical shuffling. The fiber bundle is coupled to a low-cost commercial sensor operating in GRR shutter mode. Our prototype achieves a dynamic range of up to 73dB using an 8-bit sensor with 48dB dynamic range.

eess.IV↗

arXiv2024

Can AI Extract Antecedent Factors of Human Trust in AI? An Application of Information Extraction for Scientific Literature in Behavioural and Computer Sciences

Information extraction from the scientific literature is one of the main techniques to transform unstructured knowledge hidden in the text into structured data which can then be used for decision-making in down-stream tasks. One such area is Trust in AI, where factors contributing to human trust in artificial intelligence applications are studied. The relationships of these factors with human trust in such applications are complex. We hence explore this space from the lens of information extraction where, with the input of domain experts, we carefully design annotation guidelines, create the first annotated English dataset in this domain, investigate an LLM-guided annotation, and benchmark it with state-of-the-art methods using large language models in named entity and relation extraction. Our results indicate that this problem requires supervised learning which may not be currently feasible with prompt-based LLMs.

cs.CL↗

arXiv2024

A Critical Look at Meta-evaluating Summarisation Evaluation Metrics

Effective summarisation evaluation metrics enable researchers and practitioners to compare different summarisation systems efficiently. Estimating the effectiveness of an automatic evaluation metric, termed meta-evaluation, is a critically important research question. In this position paper, we review recent meta-evaluation practices for summarisation evaluation metrics and find that (1) evaluation metrics are primarily meta-evaluated on datasets consisting of examples from news summarisation datasets, and (2) there has been a noticeable shift in research focus towards evaluating the faithfulness of generated summaries. We argue that the time is ripe to build more diverse benchmarks that enable the development of more robust evaluation metrics and analyze the generalization ability of existing evaluation metrics. In addition, we call for research focusing on user-centric quality dimensions that consider the generated summary's communicative goal and the role of summarisation in the workflow.

cs.CL↗

arXiv2024

MultiADE: A Multi-domain Benchmark for Adverse Drug Event Extraction

Active adverse event surveillance monitors Adverse Drug Events (ADE) from different data sources, such as electronic health records, medical literature, social media and search engine logs. Over the years, many datasets have been created, and shared tasks have been organised to facilitate active adverse event surveillance. However, most - if not all - datasets or shared tasks focus on extracting ADEs from a particular type of text. Domain generalisation - the ability of a machine learning model to perform well on new, unseen domains (text types) - is under-explored. Given the rapid advancements in natural language processing, one unanswered question is how far we are from having a single ADE extraction model that is effective on various types of text, such as scientific literature and social media posts. We contribute to answering this question by building a multi-domain benchmark for adverse drug event extraction, which we named MultiADE. The new benchmark comprises several existing datasets sampled from different text types and our newly created dataset - CADECv2, which is an extension of CADEC, covering online posts regarding more diverse drugs than CADEC. Our new dataset is carefully annotated by human annotators following detailed annotation guidelines. Our benchmark results show that the generalisation of the trained models is far from perfect, making it infeasible to be deployed to process different types of text. In addition, although intermediate transfer learning is a promising approach to utilising existing resources, further investigation is needed on methods of domain adaptation, particularly cost-effective methods to select useful training instances. The newly created CADECv2 and the scripts for building the benchmark are publicly available at CSIRO's Data Portal.

cs.CL↗

arXiv2024

Identifying Health Risks from Family History: A Survey of Natural Language Processing Techniques

Electronic health records include information on patients' status and medical history, which could cover the history of diseases and disorders that could be hereditary. One important use of family history information is in precision health, where the goal is to keep the population healthy with preventative measures. Natural Language Processing (NLP) and machine learning techniques can assist with identifying information that could assist health professionals in identifying health risks before a condition is developed in their later years, saving lives and reducing healthcare costs. We survey the literature on the techniques from the NLP field that have been developed to utilise digital health records to identify risks of familial diseases. We highlight that rule-based methods are heavily investigated and are still actively used for family history extraction. Still, more recent efforts have been put into building neural models based on large-scale pre-trained language models. In addition to the areas where NLP has successfully been utilised, we also identify the areas where more research is needed to unlock the value of patients' records regarding data collection, task formulation and downstream applications.

cs.CL↗

arXiv2023

Tensorial tomographic Fourier Ptychography with applications to muscle tissue imaging

We report Tensorial tomographic Fourier Ptychography (ToFu), a new non-scanning label-free tomographic microscopy method for simultaneous imaging of quantitative phase and anisotropic specimen information in 3D. Built upon Fourier Ptychography, a quantitative phase imaging technique, ToFu additionally highlights the vectorial nature of light. The imaging setup consists of a standard microscope equipped with an LED matrix, a polarization generator, and a polarization-sensitive camera. Permittivity tensors of anisotropic samples are computationally recovered from polarized intensity measurements across three dimensions. We demonstrate ToFu's efficiency through volumetric reconstructions of refractive index, birefringence, and orientation for various validation samples, as well as tissue samples from muscle fibers and diseased heart tissue. Our reconstructions of muscle fibers resolve their 3D fine-filament structure and yield consistent morphological measurements compared to gold-standard second harmonic generation scanning confocal microscope images found in the literature. Additionally, we demonstrate reconstructions of a heart tissue sample that carries important polarization information for detecting cardiac amyloidosis.

physics.optics↗

arXiv2022

Detecting Entities in the Astrophysics Literature: A Comparison of Word-based and Span-based Entity Recognition Methods

Information Extraction from scientific literature can be challenging due to the highly specialised nature of such text. We describe our entity recognition methods developed as part of the DEAL (Detecting Entities in the Astrophysics Literature) shared task. The aim of the task is to build a system that can identify Named Entities in a dataset composed by scholarly articles from astrophysics literature. We planned our participation such that it enables us to conduct an empirical comparison between word-based tagging and span-based classification methods. When evaluated on two hidden test sets provided by the organizer, our best-performing submission achieved $F_1$ scores of 0.8307 (validation phase) and 0.7990 (testing phase).

cs.CL↗

arXiv2022

Revisiting Transformer-based Models for Long Document Classification

The recent literature in text classification is biased towards short text sequences (e.g., sentences or paragraphs). In real-world applications, multi-page multi-paragraph documents are common and they cannot be efficiently encoded by vanilla Transformer-based models. We compare different Transformer-based Long Document Classification (TrLDC) approaches that aim to mitigate the computational overhead of vanilla transformers to encode much longer text, namely sparse attention and hierarchical encoding methods. We examine several aspects of sparse attention (e.g., size of local attention window, use of global attention) and hierarchical (e.g., document splitting strategy) transformers on four document classification datasets covering different domains. We observe a clear benefit from being able to process longer text, and, based on our results, we derive practical advice of applying Transformer-based models on long document classification tasks.

cs.CL↗

arXiv2022

An Exploration of Hierarchical Attention Transformers for Efficient Long Document Classification

Non-hierarchical sparse attention Transformer-based models, such as Longformer and Big Bird, are popular approaches to working with long documents. There are clear benefits to these approaches compared to the original Transformer in terms of efficiency, but Hierarchical Attention Transformer (HAT) models are a vastly understudied alternative. We develop and release fully pre-trained HAT models that use segment-wise followed by cross-segment encoders and compare them with Longformer models and partially pre-trained HATs. In several long document downstream classification tasks, our best HAT model outperforms equally-sized Longformer models while using 10-20% less GPU memory and processing documents 40-45% faster. In a series of ablation studies, we find that HATs perform best with cross-segment contextualization throughout the model than alternative configurations that implement either early or late cross-segment contextualization. Our code is on GitHub: https://github.com/coastalcph/hierarchical-transformers.

cs.CL↗

arXiv2022

Tensorial tomographic differential phase-contrast microscopy

We report Tensorial Tomographic Differential Phase-Contrast microscopy (T2DPC), a quantitative label-free tomographic imaging method for simultaneous measurement of phase and anisotropy. T2DPC extends differential phase-contrast microscopy, a quantitative phase imaging technique, to highlight the vectorial nature of light. The method solves for permittivity tensor of anisotropic samples from intensity measurements acquired with a standard microscope equipped with an LED matrix, a circular polarizer, and a polarization-sensitive camera. We demonstrate accurate volumetric reconstructions of refractive index, birefringence, and orientation for various validation samples, and show that the reconstructed polarization structures of a biological specimen are predictive of pathology.

physics.optics↗

arXiv2021

Quantitative Jones matrix imaging using vectorial Fourier ptychography

This paper presents a microscopic imaging technique that uses variable-angle illumination to recover the complex polarimetric properties of a specimen at high resolution and over a large field-of-view. The approach extends Fourier ptychography, which is a synthetic aperture-based imaging approach to improve resolution with phaseless measurements, to additionally account for the vectorial nature of light. After images are acquired using a standard microscope outfitted with an LED illumination array and two polarizers, our vectorial Fourier Ptychography (vFP) algorithm solves for the complex 2x2 Jones matrix of the anisotropic specimen of interest at each resolved spatial location. We introduce a new sequential Gauss-Newton-based solver that additionally jointly estimates and removes polarization-dependent imaging system aberrations. We demonstrate effective vFP performance by generating large-area (29 mm$^2$), high-resolution (1.24 $μ$m full-pitch) reconstructions of sample absorption, phase, orientation, diattenuation, and retardance for a variety of calibration samples and biological specimens.

physics.optics↗

arXiv2021

MDAPT: Multilingual Domain Adaptive Pretraining in a Single Model

Domain adaptive pretraining, i.e. the continued unsupervised pretraining of a language model on domain-specific text, improves the modelling of text for downstream tasks within the domain. Numerous real-world applications are based on domain-specific text, e.g. working with financial or biomedical documents, and these applications often need to support multiple languages. However, large-scale domain-specific multilingual pretraining data for such scenarios can be difficult to obtain, due to regulations, legislation, or simply a lack of language- and domain-specific text. One solution is to train a single multilingual model, taking advantage of the data available in as many languages as possible. In this work, we explore the benefits of domain adaptive pretraining with a focus on adapting to multiple languages within a specific domain. We propose different techniques to compose pretraining corpora that enable a language model to both become domain-specific and multilingual. Evaluation on nine domain-specific datasets-for biomedical named entity recognition and financial sentence classification-covering seven different languages show that a single multilingual domain-specific model can outperform the general multilingual model, and performs close to its monolingual counterpart. This finding holds across two different pretraining methods, adapter-based pretraining and full model pretraining.

cs.CL↗

arXiv2021

Recognising Biomedical Names: Challenges and Solutions

The growth rate in the amount of biomedical documents is staggering. Unlocking information trapped in these documents can enable researchers and practitioners to operate confidently in the information world. Biomedical NER, the task of recognising biomedical names, is usually employed as the first step of the NLP pipeline. Standard NER models, based on sequence tagging technique, are good at recognising short entity mentions in the generic domain. However, there are several open challenges of applying these models to recognise biomedical names: 1) Biomedical names may contain complex inner structure (discontinuity and overlapping) which cannot be recognised using standard sequence tagging technique; 2) The training of NER models usually requires large amount of labelled data, which are difficult to obtain in the biomedical domain; and, 3) Commonly used language representation models are pre-trained on generic data; a domain shift therefore exists between these models and target biomedical data. To deal with these challenges, we explore several research directions and make the following contributions: 1) we propose a transition-based NER model which can recognise discontinuous mentions; 2) We develop a cost-effective approach that nominates the suitable pre-training data; and, 3) We design several data augmentation methods for NER. Our contributions have obvious practical implications, especially when new biomedical applications are needed. Our proposed data augmentation methods can help the NER model achieve decent performance, requiring only a small amount of labelled data. Our investigation regarding selecting pre-training data can improve the model by incorporating language representation models, which are pre-trained using in-domain data. Finally, our proposed transition-based NER model can further improve the performance by recognising discontinuous mentions.

cs.CL↗

arXiv2020

NLNDE at CANTEMIST: Neural Sequence Labeling and Parsing Approaches for Clinical Concept Extraction

The recognition and normalization of clinical information, such as tumor morphology mentions, is an important, but complex process consisting of multiple subtasks. In this paper, we describe our system for the CANTEMIST shared task, which is able to extract, normalize and rank ICD codes from Spanish electronic health records using neural sequence labeling and parsing approaches with context-aware embeddings. Our best system achieves 85.3 F1, 76.7 F1, and 77.0 MAP for the three tasks, respectively.

cs.CL↗