Search

This study focuses on the application of the Heston model to option pricing, employing both theoretical derivations and empirical validations. The Heston model, known for its ability to incorporate stochastic volatility, is derived and analyzed to evaluate its effectiveness in pricing options. For practical application, we utilize Monte Carlo simulations alongside market data from the Crude Oil WTI market to test the model's accuracy. Machine learning based optimization methods are also applied for the estimation of the five Heston parameters. By calibrating the model with real-world data, we assess its robustness and relevance in current financial markets, aiming to bridge the gap between theoretical finance models and their practical implementations., Comment: 32 pages, 18 figures, 3 tables

This thesis explores the historical progression and theoretical constructs of financial mathematics, with an in-depth exploration of Stochastic Calculus as showcased in the Binomial Asset Pricing Model and the Continuous-Time Models. A comprehensive survey of stochastic calculus principles applied to option pricing is offered, highlighting insights from Peter Carr and Lorenzo Torricelli's ``Convex Duality in Continuous Option Pricing Models". This manuscript adopts techniques such as Monte-Carlo Simulation and machine learning algorithms to examine the propositions of Carr and Torricelli, drawing comparisons between the Logistic and Bachelier models. Additionally, it suggests directions for potential future research on option pricing methods., Comment: 65 pages, 17 figures, 6 tables

Tracking in gigapixel scenarios holds numerous potential applications in video surveillance and pedestrian analysis. Existing algorithms attempt to perform tracking in crowded scenes by utilizing multiple cameras or group relationships. However, their performance significantly degrades when confronted with complex interaction and occlusion inherent in gigapixel images. In this paper, we introduce DynamicTrack, a dynamic tracking framework designed to address gigapixel tracking challenges in crowded scenes. In particular, we propose a dynamic detector that utilizes contrastive learning to jointly detect the head and body of pedestrians. Building upon this, we design a dynamic association algorithm that effectively utilizes head and body information for matching purposes. Extensive experiments show that our tracker achieves state-of-the-art performance on widely used tracking benchmarks specifically designed for gigapixel crowded scenes.

URLs play a crucial role in understanding and categorizing web content, particularly in tasks related to security control and online recommendations. While pre-trained models are currently dominating various fields, the domain of URL analysis still lacks specialized pre-trained models. To address this gap, this paper introduces URLBERT, the first pre-trained representation learning model applied to a variety of URL classification or detection tasks. We first train a URL tokenizer on a corpus of billions of URLs to address URL data tokenization. Additionally, we propose two novel pre-training tasks: (1) self-supervised contrastive learning tasks, which strengthen the model's understanding of URL structure and the capture of category differences by distinguishing different variants of the same URL; (2) virtual adversarial training, aimed at improving the model's robustness in extracting semantic features from URLs. Finally, our proposed methods are evaluated on tasks including phishing URL detection, web page classification, and ad filtering, achieving state-of-the-art performance. Importantly, we also explore multi-task learning with URLBERT, and experimental results demonstrate that multi-task learning model based on URLBERT exhibit equivalent effectiveness compared to independently fine-tuned models, showing the simplicity of URLBERT in handling complex task requirements. The code for our work is available at https://github.com/Davidup1/URLBERT.

Astigmatic unitary transformations allow for the adiabatic connections of all feasible states of paraxial Gaussian beams on the same modal sphere, i.e., Hermite-Laguerre-Gaussian (HLG) modes. Here, we present a comprehensive investigation into the unitary modal evolution of complex structured Gaussian beams, comprised by HLG modes from disparate modal spheres, via astigmatic transformation. The non-synchronized higher-order geometric phases in cyclic transformations originates a Talbot-effect-like modal evolution in the superposition state of these HLG modes, resulting in pattern variations and revivals in transformations with specific geodesic loops. Using Ince-Gaussian modes as an illustrative example, we systematically analyze and experimentally corroborate the beamforming mechanism behind the pattern evolution. Our results outline a generic modal conversion theory of structured Gaussian beams via astigmatic unitary transformation, offering a new approach for shaping spatial modal structure. These findings may inspire a wide variety of applications based on structured light., Comment: 7 pages, 5 figures

As blockchain smart contracts become more widespread and carry more valuable digital assets, they become an increasingly attractive target for attackers. Over the past few years, smart contracts have been subject to a plethora of devastating attacks, resulting in billions of dollars in financial losses. There has been a notable surge of research interest in identifying defects in smart contracts. However, existing smart contract fuzzing tools are still unsatisfactory. They struggle to screen out meaningful transaction sequences and specify critical inputs for each transaction. As a result, they can only trigger a limited range of contract states, making it difficult to unveil complicated vulnerabilities hidden in the deep state space. In this paper, we shed light on smart contract fuzzing by employing a sequence-aware mutation and seed mask guidance strategy. In particular, we first utilize data-flow-based feedback to determine transaction orders in a meaningful way and further introduce a sequence-aware mutation technique to explore deeper states. Thereafter, we design a mask-guided seed mutation strategy that biases the generated transaction inputs to hit target branches. In addition, we develop a dynamic-adaptive energy adjustment paradigm that balances the fuzzing resource allocation during a fuzzing campaign. We implement our designs into a new smart contract fuzzer named MuFuzz, and extensively evaluate it on three benchmarks. Empirical results demonstrate that MuFuzz outperforms existing tools in terms of both branch coverage and bug finding. Overall, MuFuzz achieves higher branch coverage than state-of-the-art fuzzers (up to 25%) and detects 30% more bugs than existing bug detectors., Comment: This paper has been accepted by ICDE 2024

The widespread use of the Internet has revolutionized information retrieval methods. However, this transformation has also given rise to a significant cybersecurity challenge: the rapid proliferation of malicious URLs, which serve as entry points for a wide range of cyber threats. In this study, we present an efficient pre-training model-based framework for malicious URL detection. Leveraging the subword and character-aware pre-trained model, CharBERT, as our foundation, we further develop three key modules: hierarchical feature extraction, layer-aware attention, and spatial pyramid pooling. The hierarchical feature extraction module follows the pyramid feature learning principle, extracting multi-level URL embeddings from the different Transformer layers of CharBERT. Subsequently, the layer-aware attention module autonomously learns connections among features at various hierarchical levels and allocates varying weight coefficients to each level of features. Finally, the spatial pyramid pooling module performs multiscale downsampling on the weighted multi-level feature pyramid, achieving the capture of local features as well as the aggregation of global features. The proposed method has been extensively validated on multiple public datasets, demonstrating a significant improvement over prior works, with the maximum accuracy gap reaching 8.43% compared to the previous state-of-the-art method. Additionally, we have assessed the model's generalization and robustness in scenarios such as cross-dataset evaluation and adversarial attacks. Finally, we conducted real-world case studies on the active phishing URLs., Comment: 11 pages, 7 figures

Neutron stars (NSs) which could contain exotic degrees of freedom in the core and the self-bound quark stars (QSs) made purely of absolutely stable deconfined quark matter are still two main candidates for the compact objects observed in pulsars and gravitational wave (GW) events in binary star mergers. We perform a Bayesian model-agnostic inference of the properties of NSs and QSs by combining multi-messenger data of GW170817, GW190425, PSR J0030+0451, PSR J0740+6620, PSR J1614-2230, PSR J0348+0432 as well as ab initio calculations from perturbative quantum chromodynamics and chiral effective field theory. We find the NS scenario is strongly favored against the QS scenario with a Bayes factor of NS over QS $\mathcal{B}^\text{NS}_\text{QS} = 11.5$. In addition, the peak of the squared sound velocity $c_s^2 \sim 0.5c^2$ around $3.5$ times nuclear saturation density $n_0$ observed in the NS case disappears in the QS case which suggests that the $c_s^2$ first increases and then saturates at $c_s^2 \sim 0.5c^2$ above $\sim 4n_0$. The sound velocity and trace anomaly are found to approach the conformal limit in the core of heavy NSs with mass $M \gtrsim 2M_{\odot}$, but not in the core of QSs., Comment: 6 pages, 2 figures

This paper introduced key aspects of applying Machine Learning (ML) models, improved trading strategies, and the Quasi-Reversibility Method (QRM) to optimize stock option forecasting and trading results. It presented the findings of the follow-up project of the research "Application of Convolutional Neural Networks with Quasi-Reversibility Method Results for Option Forecasting". First, the project included an application of Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) networks to provide a novel way of predicting stock option trends. Additionally, it examined the dependence of the ML models by evaluating the experimental method of combining multiple ML models to improve prediction results and decision-making. Lastly, two improved trading strategies and simulated investing results were presented. The Binomial Asset Pricing Model with discrete time stochastic process analysis and portfolio hedging was applied and suggested an optimized investment expectation. These results can be utilized in real-life trading strategies to optimize stock option investment results based on historical data., Comment: 11 pages

Practical dialog systems need to deal with various knowledge sources, noisy user expressions, and the shortage of annotated data. To better solve the above problems, we propose CGoDial, new challenging and comprehensive Chinese benchmark for multi-domain Goal-oriented Dialog evaluation. It contains 96,763 dialog sessions and 574,949 dialog turns totally, covering three datasets with different knowledge sources: 1) a slot-based dialog (SBD) dataset with table-formed knowledge, 2) a flow-based dialog (FBD) dataset with tree-formed knowledge, and a retrieval-based dialog (RBD) dataset with candidate-formed knowledge. To bridge the gap between academic benchmarks and spoken dialog scenarios, we either collect data from real conversations or add spoken features to existing datasets via crowd-sourcing. The proposed experimental settings include the combinations of training with either the entire training set or a few-shot training set, and testing with either the standard test set or a hard test subset, which can assess model capabilities in terms of general prediction, fast adaptability and reliable robustness., Comment: EMNLP 2022

In this paper, we conduct systematic measurement studies to show that the high memory bandwidth consumption of modern distributed applications can lead to a significant drop of network throughput and a large increase of tail latency in high-speed RDMA networks.We identify its root cause as the high contention of memory bandwidth between application processes and network processes. This contention leads to frequent packet drops at the NIC of receiving hosts, which triggers the congestion control mechanism of the network and eventually results in network performance degradation. To tackle this problem, we make a key observation that given the distributed storage service, the vast majority of data it receives from the network will be eventually written to high-speed storage media (e.g., SSD) by CPU. As such, we propose to bypass host memory when processing received data to completely circumvent this performance bottleneck. In particular, we design Lamda, a novel receiver cache processing system that consumes a small amount of CPU cache to process received data from the network at line rate. We implement a prototype of Lamda and evaluate its performance extensively in a Clos-based testbed. Results show that for distributed storage applications, Lamda improves network throughput by 4.7% with zero memory bandwidth consumption on storage nodes, and improves network throughput by up 17% and 45% for large block size and small size under the memory bandwidth pressure, respectively. Lamda can also be applied to latency-sensitive HPC applications, which reduces their communication latency by 35.1%.

Deep learning has shown promise in enhancing channel state information (CSI) feedback. However, many studies indicate that better feedback performance often accompanies higher computational complexity. Pursuing better performance-complexity tradeoffs is crucial to facilitate practical deployment, especially on computation-limited devices, which may have to use lightweight autoencoder with unfavorable performance. To achieve this goal, this paper introduces knowledge distillation (KD) to achieve better tradeoffs, where knowledge from a complicated teacher autoencoder is transferred to a lightweight student autoencoder for performance improvement. Specifically, two methods are proposed for implementation. Firstly, an autoencoder KD-based method is introduced by training a student autoencoder to mimic the reconstructed CSI of a pretrained teacher autoencoder. Secondly, an encoder KD-based method is proposed to reduce training overhead by performing KD only on the student encoder. Additionally, a variant of encoder KD is introduced to protect user equipment and base station vendor intellectual property. Numerical simulations demonstrate that the proposed methods can significantly improve the student autoencoder's performance, while reducing the number of floating point operations and inference time to 3.05%-5.28% and 13.80%-14.76% of the teacher network, respectively. Furthermore, the variant encoder KD method effectively enhances the student autoencoder's generalization capability across different scenarios, environments, and bandwidths., Comment: 13 pages, 5 figures

In this paper, we propose a novel SQL guided pre-training framework STAR for context-dependent text-to-SQL parsing, which leverages contextual information to enrich natural language (NL) utterance and table schema representations for text-to-SQL conversations. Concretely, we propose two novel pre-training objectives which respectively explore the context-dependent interactions of NL utterances and SQL queries within each text-to-SQL conversation: (i) schema state tracking (SST) objective that tracks and explores the schema states of context-dependent SQL queries in the form of schema-states by predicting and updating the value of each schema slot during interaction; (ii) utterance dependency tracking (UDT) objective that employs weighted contrastive learning to pull together two semantically similar NL utterances and push away the representations of semantically dissimilar NL utterances within each conversation. In addition, we construct a high-quality large-scale context-dependent text-to-SQL conversation corpus to pre-train STAR. Extensive experiments show that STAR achieves new state-of-the-art performance on two downstream benchmarks (SParC and CoSQL), significantly outperforming previous pre-training methods and ranking first on the leaderboard. We believe the release of the constructed corpus, codebase and pre-trained STAR checkpoints would push forward the research in this area. For reproducibility, we release our code and data at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/star., Comment: EMNLP 2022

This paper presents a methodology for combining programming and mathematics to optimize elevator wait times. Based on simulated user data generated according to the canonical three-peak model of elevator traffic, we first develop a naive model from an intuitive understanding of the logic behind elevators. We take into consideration a general array of features including capacity, acceleration, and maximum wait time thresholds to adequately model realistic circumstances. Using the same evaluation framework, we proceed to develop a Deep Q Learning model in an attempt to match the hard-coded naive approach for elevator control. Throughout the majority of the paper, we work under a Markov Decision Process (MDP) schema, but later explore how the assumption fails to characterize the highly stochastic overall Elevator Group Control System (EGCS)., Comment: 16 pages

This research investigates analytical and quantitative methods for simulating elevator optimizations. To maximize overall elevator usage, we concentrate on creating a multiple-user positive-sum system that is inspired by agent-based game theory. We define and create basic "Dumbwaiter" models by attempting both the Spatial Process Approach and the Gibbs Random Field Approach. These two mathematical techniques approach the problem from different points of view: the spatial process can give an analytical solution in continuous space and the Gibbs Random Field provides a discrete framework to flexibly model the problem on a computer. Starting from the simplest case, we target the assumptions to provide concrete solutions to the models and develop a "Multi-Dumbwaiter System". This paper examines, evaluates, and proves the ultimate success of such implemented strategies to design the basic elevator's optimal policy; consequently, not only do we believe in the results' practicality for industry, but also their potential for application., Comment: 14 pages

This paper outlines two approaches for mathematical, simulation, modeling, and analysis of hypothetical creatures, in particular, the dragons of HBO's television series Game of Thrones (GOT). Our first approach, the forward model, utilizes quasi-empirical observations of various features of GOT dragons. We then mathematically derive the growth rate, other dimensions, energy consumption, etc. In the backward model, we use projected energy consumption by given ecological impact to model an expected dragon in terms of physical features. We compare and contrast both models to examine the plausibility of a real-world existence for our titular dragons and provide brief analyses of potential impacts on ecology., Comment: 16 pages

Pre-training methods with contrastive learning objectives have shown remarkable success in dialog understanding tasks. However, current contrastive learning solely considers the self-augmented dialog samples as positive samples and treats all other dialog samples as negative ones, which enforces dissimilar representations even for dialogs that are semantically related. In this paper, we propose SPACE-2, a tree-structured pre-trained conversation model, which learns dialog representations from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised contrastive pre-training. Concretely, we first define a general semantic tree structure (STS) to unify the inconsistent annotation schema across different dialog datasets, so that the rich structural information stored in all labeled data can be exploited. Then we propose a novel multi-view score function to increase the relevance of all possible dialogs that share similar STSs and only push away other completely different dialogs during supervised contrastive pre-training. To fully exploit unlabeled dialogs, a basic self-supervised contrastive loss is also added to refine the learned representations. Experiments show that our method can achieve new state-of-the-art results on the DialoGLUE benchmark consisting of seven datasets and four popular dialog understanding tasks. For reproducibility, we release the code and data at https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/space-2., Comment: 17 pages, 6 figures. Accepted by COLING 2022

Deep neural networks trained with standard cross-entropy loss are more prone to memorize noisy labels, which degrades their performance. Negative learning using complementary labels is more robust when noisy labels intervene but with an extremely slow model convergence speed. In this paper, we first introduce a bidirectional learning scheme, where positive learning ensures convergence speed while negative learning robustly copes with label noise. Further, a dynamic sample reweighting strategy is proposed to globally weaken the effect of noise-labeled samples by exploiting the excellent discriminatory ability of negative learning on the sample probability distribution. In addition, we combine self-distillation to further improve the model performance. The code is available at \url{https://github.com/chenchenzong/BLDR}.

This paper presents a novel way to apply mathematical finance and machine learning (ML) to forecast stock options prices. Following results from the paper Quasi-Reversibility Method and Neural Network Machine Learning to Solution of Black-Scholes Equations (appeared on the AMS Contemporary Mathematics journal), we create and evaluate new empirical mathematical models for the Black-Scholes equation to analyze data for 92,846 companies. We solve the Black-Scholes (BS) equation forwards in time as an ill-posed inverse problem, using the Quasi-Reversibility Method (QRM), to predict option price for the future one day. For each company, we have 13 elements including stock and option daily prices, volatility, minimizer, etc. Because the market is so complicated that there exists no perfect model, we apply ML to train algorithms to make the best prediction. The current stage of research combines QRM with Convolutional Neural Networks (CNN), which learn information across a large number of data points simultaneously. We implement CNN to generate new results by validating and testing on sample market data. We test different ways of applying CNN and compare our CNN models with previous models to see if achieving a higher profit rate is possible., Comment: 10 pages, 2 figures, 5 tables

The emergence of metaverse brings tremendous evolution to Non-Fungible Tokens (NFTs), which could certify the ownership the unique digital asset in the cyber world. The NFT market has garnered unprecedented attention from investors and created billions of dollars in transaction volume. Meanwhile, securing NFT is still a challenging issue. Recently, numerous incidents of NFT theft have been reported, leading to incalculable losses for holders. We propose a decentralized NFT anti-theft mechanism called TokenPatronus, which supports the general ERC-721 standard and provide the holders with strong property protection. TokenPatronus contains pre-event protection, in-event interruption, and post-event replevin enhancements for the complete NFTs transactions stages. Four modules are designed to make up the decentralized anti-theft mechanism, including the decentralized access control (DAC), the decentralized risk management (DRM), the decentralized arbitration system (DAS) and the ERC-721G standard smart contract. TokenPatronus is performing on the Turtlecase NFT project of Ethereum and will support more blockchains in the future., Comment: submitted to CESC 2022 as a work-in-progress paper

Large-scale pretrained foundation models have been an emerging paradigm for building artificial intelligence (AI) systems, which can be quickly adapted to a wide range of downstream tasks. This paper presents mPLUG, a new vision-language foundation model for both cross-modal understanding and generation. Most existing pre-trained models suffer from the problems of low computational efficiency and information asymmetry brought by the long visual sequence in cross-modal alignment. To address these problems, mPLUG introduces an effective and efficient vision-language architecture with novel cross-modal skip-connections, which creates inter-layer shortcuts that skip a certain number of layers for time-consuming full self-attention on the vision side. mPLUG is pre-trained end-to-end on large-scale image-text pairs with both discriminative and generative objectives. It achieves state-of-the-art results on a wide range of vision-language downstream tasks, such as image captioning, image-text retrieval, visual grounding and visual question answering. mPLUG also demonstrates strong zero-shot transferability when directly transferred to multiple video-language tasks.

We develop a relativistic (quasi-)hydrodynamic framework, dubbed the gyrohydrodynamics, to describe fluid dynamics of many-body systems with spin under strong vorticity based on entropy-current analysis. This framework generalizes the recently-developed spin hydrodynamics to the regime where the spin density is at the leading order in derivatives but suppressed by another small parameter, the Planck constant $\hbar$, due to its quantum nature. Our analysis shows that the complete first-order constitutive relations of gyrohydrodynamics involve seventeen transport coefficients and are highly anisotropic.

Deep learning has been used in a wide range of areas and made a huge breakthrough. With the ever-increasing model size and train-ing data volume, distributed deep learning emerges which utilizes a cluster to train a model in parallel. Unfortunately, the performance is often far from linear speedup due to the communication overhead between cluster nodes. To address this challenge, this paper designs and implements Libra, a network aggregator, that utilizes in-network computation to optimize the communication for distributed DL training in two aspects: 1) reduce active connections and 2) aggregate exchanged network packets. We implemented our Libra on Intel Tofino switches, customized a lightweight host stack and integrated it into an open-source training framework PS-lite. The experimental result shows that our Libra can achieve 1.5~4 times speedup., Comment: 13 pages, 18 figures

We present a tabulated version of our slim disk model for fitting tidal disruption events (TDEs). We create a synthetic X-ray spectral library by ray-tracing stationary general relativistic slim disks and including gravitational redshift, Doppler, and lensing effects self-consistently. We introduce the library to reduce computational expense and increase access for fitting future events. Fitting requires interpolation between the library spectra; the interpolation error in the synthetic flux is generally $<10\%$ (it can rise to $40\%$ when the disk is nearly edge-on). We fit the X-ray spectra of the TDEs ASASSN-14li and ASASSN-15oi, successfully reproducing our earlier constraints on black hole mass $M_\bullet$ and spin $a_\bullet$ from full on-the-fly ray-tracing. We use the library to fit mock observational data to explore the degeneracies among parameters, finding that 1) hotter thermal disk and edge-on inclination angle spectra offer tighter constraints on $M_\bullet$ and $a_\bullet$; 2) the constraining power of spectra on $M_\bullet$ and $a_\bullet$ increases as a power-law with the number of X-ray counts, and the index of the power law is higher for hotter thermal disk spectra; 3) multi-epoch X-ray spectra partially break the degeneracy between $M_\bullet$ and $a_\bullet$; 4) the time-dependent level of X-ray absorption can be constrained from spectral fitting. The tabulated model and slim disk model are {\href{https://doi.org/10.25739/hfhz-xn60}{here.}, Comment: Accepted for Publication in ApJ

Pre-trained models have proved to be powerful in enhancing task-oriented dialog systems. However, current pre-training methods mainly focus on enhancing dialog understanding and generation tasks while neglecting the exploitation of dialog policy. In this paper, we propose GALAXY, a novel pre-trained dialog model that explicitly learns dialog policy from limited labeled dialogs and large-scale unlabeled dialog corpora via semi-supervised learning. Specifically, we introduce a dialog act prediction task for policy optimization during pre-training and employ a consistency regularization term to refine the learned representation with the help of unlabeled dialogs. We also implement a gating mechanism to weigh suitable unlabeled dialog samples. Empirical results show that GALAXY substantially improves the performance of task-oriented dialog systems, and achieves new state-of-the-art results on benchmark datasets: In-Car, MultiWOZ2.0 and MultiWOZ2.1, improving their end-to-end combined scores by 2.5, 5.3 and 5.5 points, respectively. We also show that GALAXY has a stronger few-shot ability than existing models under various low-resource settings., Comment: 7 pages, 5 figures. Accepted by AAAI 2022

Recently pre-training models have significantly improved the performance of various NLP tasks by leveraging large-scale text corpora to improve the contextual representation ability of the neural network. The large pre-training language model has also been applied in the area of table semantic parsing. However, existing pre-training approaches have not carefully explored explicit interaction relationships between a question and the corresponding database schema, which is a key ingredient for uncovering their semantic and structural correspondence. Furthermore, the question-aware representation learning in the schema grounding context has received less attention in pre-training objective.To alleviate these issues, this paper designs two novel pre-training objectives to impose the desired inductive bias into the learned representations for table pre-training. We further propose a schema-aware curriculum learning approach to mitigate the impact of noise and learn effectively from the pre-training data in an easy-to-hard manner. We evaluate our pre-trained framework by fine-tuning it on two benchmarks, Spider and SQUALL. The results demonstrate the effectiveness of our pre-training objective and curriculum compared to a variety of baselines.

The Visual Question Answering (VQA) task utilizes both visual image and language analysis to answer a textual question with respect to an image. It has been a popular research topic with an increasing number of real-world applications in the last decade. This paper describes our recent research of AliceMind-MMU (ALIbaba's Collection of Encoder-decoders from Machine IntelligeNce lab of Damo academy - MultiMedia Understanding) that obtains similar or even slightly better results than human being does on VQA. This is achieved by systematically improving the VQA pipeline including: (1) pre-training with comprehensive visual and textual feature representation; (2) effective cross-modal interaction with learning to attend; and (3) A novel knowledge mining framework with specialized expert modules for the complex VQA task. Treating different types of visual questions with corresponding expertise needed plays an important role in boosting the performance of our VQA architecture up to the human level. An extensive set of experiments and analysis are conducted to demonstrate the effectiveness of the new research work.

Galactic black hole X-ray binaries (BHXBs) provide excellent laboratories to study accretion, as their relatively quick evolution allows us to monitor large changes in the in-flowing and/or out-flowing material over human timescales. However, the details of how the inflow-outflow coupling evolves during a BHXB outburst remain an area of active debate. In this work we attempt to probe the physical changes underlying the system evolution, by performing a systematic analysis of the multi-wavelength data of three BHXB sources: XTE J1752-223, MAXI J1659-152, and XTE J1650-500, during hard and hard-intermediate states. Using the power spectral hue which characterises the X-ray variability properties, we identify several clusters of BHXB epochs and perform the joint multi-wavelength spectral modelling to test their commonality with a physical jet model. Under the assumption that the corona is related to the base of the jet, we find that the power spectral hue traces the variation of the coronal radius(from ~10Rg - ~40Rg) in multiple BHXBs at hard and hard-intermediate states, and that the data are consistent with moderately truncated accretion discs (<25Rg) during hard-intermediate states. We also find that all epochs of low disc reflection have high hue located near the hard-intermediate to soft-intermediate state transition, indicating that in these states the vertical extent of the corona and/or its bulk speed are increasing. Our results link the geometrical similarity in the corona among multiple BHXB sources to their timing characteristics, and probe the corona responding to the disc-jet interactions at hard and intermediate states during outbursts., Comment: 12 pages, 10 figures. Accepted for publication in MNRAS