Search

Large Vision-Language Models (LVLMs), trained on multimodal big datasets, have significantly advanced AI by excelling in vision-language tasks. However, these models remain vulnerable to adversarial attacks, particularly jailbreak attacks, which bypass safety protocols and cause the model to generate misleading or harmful responses. This vulnerability stems from both the inherent susceptibilities of LLMs and the expanded attack surface introduced by the visual modality. We propose Sim-CLIP+, a novel defense mechanism that adversarially fine-tunes the CLIP vision encoder by leveraging a Siamese architecture. This approach maximizes cosine similarity between perturbed and clean samples, facilitating resilience against adversarial manipulations. Sim-CLIP+ offers a plug-and-play solution, allowing seamless integration into existing LVLM architectures as a robust vision encoder. Unlike previous defenses, our method requires no structural modifications to the LVLM and incurs minimal computational overhead. Sim-CLIP+ demonstrates effectiveness against both gradient-based adversarial attacks and various jailbreak techniques. We evaluate Sim-CLIP+ against three distinct jailbreak attack strategies and perform clean evaluations using standard downstream datasets, including COCO for image captioning and OKVQA for visual question answering. Extensive experiments demonstrate that Sim-CLIP+ maintains high clean accuracy while substantially improving robustness against both gradient-based adversarial attacks and jailbreak techniques. Our code and robust vision encoders are available at https://github.com/speedlab-git/Robust-Encoder-against-Jailbreak-attack.git.

The rapid advancement and increasing complexity of pretrained models, exemplified by CLIP, offer significant opportunities as well as challenges for Federated Learning (FL), a critical component of privacy-preserving artificial intelligence. This research delves into the intricacies of integrating large foundation models like CLIP within FL frameworks to enhance privacy, efficiency, and adaptability across heterogeneous data landscapes. It specifically addresses the challenges posed by non-IID data distributions, the computational and communication overheads of leveraging such complex models, and the skewed representation of classes within datasets. We propose TriplePlay, a framework that integrates CLIP as an adapter to enhance FL's adaptability and performance across diverse data distributions. This approach addresses the long-tail distribution challenge to ensure fairness while reducing resource demands through quantization and low-rank adaptation techniques.Our simulation results demonstrate that TriplePlay effectively decreases GPU usage costs and speeds up the learning process, achieving convergence with reduced communication overhead.

Hand shadow puppetry, also known as shadowgraphy or ombromanie, is a form of theatrical art and storytelling where hand shadows are projected onto flat surfaces to create illusions of living creatures. The skilled performers create these silhouettes by hand positioning, finger movements, and dexterous gestures to resemble shadows of animals and objects. Due to the lack of practitioners and a seismic shift in people's entertainment standards, this art form is on the verge of extinction. To facilitate its preservation and proliferate it to a wider audience, we introduce ${\rm H{\small A}SP{\small E}R}$, a novel dataset consisting of 8,340 images of hand shadow puppets across 11 classes extracted from both professional and amateur hand shadow puppeteer clips. We provide a detailed statistical analysis of the dataset and employ a range of pretrained image classification models to establish baselines. Our findings show a substantial performance superiority of traditional convolutional models over attention-based transformer architectures. We also find that lightweight models, such as MobileNetV2, suited for mobile applications and embedded devices, perform comparatively well. We surmise that such low-latency architectures can be useful in developing ombromanie teaching tools, and we create a prototype application to explore this surmission. Keeping the best-performing model InceptionV3 under the limelight, we conduct comprehensive feature-spatial, explainability, and error analyses to gain insights into its decision-making process. To the best of our knowledge, this is the first documented dataset and research endeavor to preserve this dying art for future generations, with computer vision approaches. Our code and data are publicly available., Comment: Submitted to IEEE Transactions on Artificial Intelligence (IEEE TAI), 11 pages, 78 figures, 2 tables

Vision-language models (VLMs) have achieved significant strides in recent times specially in multimodal tasks, yet they remain susceptible to adversarial attacks on their vision components. To address this, we propose Sim-CLIP, an unsupervised adversarial fine-tuning method that enhances the robustness of the widely-used CLIP vision encoder against such attacks while maintaining semantic richness and specificity. By employing a Siamese architecture with cosine similarity loss, Sim-CLIP learns semantically meaningful and attack-resilient visual representations without requiring large batch sizes or momentum encoders. Our results demonstrate that VLMs enhanced with Sim-CLIP's fine-tuned CLIP encoder exhibit significantly enhanced robustness against adversarial attacks, while preserving semantic meaning of the perturbed images. Notably, Sim-CLIP does not require additional training or fine-tuning of the VLM itself; replacing the original vision encoder with our fine-tuned Sim-CLIP suffices to provide robustness. This work underscores the significance of reinforcing foundational models like CLIP to safeguard the reliability of downstream VLM applications, paving the way for more secure and effective multimodal systems.

Fast Radio Bursts (FRBs), like pulsars, display radio emission from compact regions such that they can be treated as point sources. As this radiation propagates through space, they encounter sources of lensing such as a gravitational field of massive objects or inhomogeneous changes in the electron density of cold plasma. We have developed a simulation tool to generate these lensing morphologies through coherent propagation transfer functions generated by phase coherent geometric optics on a spatial grid. In the limit an FRB can be treated as a point source, the ray paths from the FRB to the observer are phase coherent. Each image will have a time delay and magnification that will alter the emitted frequency-temporal morphology of the FRB to that which is observed. The interference of these images could also decohere the observed phase properties of the images, affecting any phase related searches such as searching for the auto-correlation of the observed FRB voltage with other images in time. We present analytic test cases to demonstrate that the simulation can model qualitative properties. We provide example multi-plane lensing systems to show the capabilities of the simulation in modeling the lensed morphology of an FRB and observed phase coherence., Comment: 21 pages, 15 figures

This study introduces a novel approach to simulating legislative processes using LLM-driven virtual agents, focusing on the U.S. Senate Intelligence Committee. We developed agents representing individual senators and placed them in simulated committee discussions. The agents demonstrated the ability to engage in realistic debate, provide thoughtful reflections, and find bipartisan solutions under certain conditions. Notably, the simulation also showed promise in modeling shifts towards bipartisanship in response to external perturbations. Our results indicate that this LLM-driven approach could become a valuable tool for understanding and potentially improving legislative processes, supporting a broader pattern of findings highlighting how LLM-based agents can usefully model real-world phenomena. Future works will focus on enhancing agent complexity, expanding the simulation scope, and exploring applications in policy testing and negotiation.

Fast radio bursts (FRBs) are micro-to-millisecond duration radio transients that originate mostly from extragalactic distances. The emission mechanism responsible for these high luminosity, short duration transients remains debated. The models are broadly grouped into two classes: physical processes that occur within close proximity to a central engine; and central engines that release energy which moves to large radial distances and subsequently interacts with surrounding media producing radio waves. The expected emission region sizes are notably different between these two types of models. FRB emission size constraints can therefore be used to distinguish between these competing models and inform on the physics responsible. Here we present the measurement of two mutually coherent scintillation scales in the frequency spectrum of FRB 20221022A: one originating from a scattering screen located within the Milky Way, and the second originating from a scattering screen located within its host galaxy or local environment. We use the scattering media as an astrophysical lens to constrain the size of the lateral emission region, $R_{\star\mathrm{obs}} \lesssim 3\times10^{4}$ km. We find that this is inconsistent with the expected emission sizes for the large radial distance models, and is more naturally explained with an emission process that operates within or just beyond the magnetosphere of a central compact object. Recently, FRB 20221022A was found to exhibit an S-shaped polarisation angle swing, supporting a magnetospheric emission process. The scintillation results presented in this work independently support this conclusion, while highlighting scintillation as a useful tool in our understanding of FRB emission physics and progenitors., Comment: Submitted, comments welcome

Nuclei instance segmentation is crucial in oncological diagnosis and cancer pathology research. H&E stained images are commonly used for medical diagnosis, but pre-processing is necessary before using them for image processing tasks. Two principal pre-processing methods are formalin-fixed paraffin-embedded samples (FFPE) and frozen tissue samples (FS). While FFPE is widely used, it is time-consuming, while FS samples can be processed quickly. Analyzing H&E stained images derived from fast sample preparation, staining, and scanning can pose difficulties due to the swift process, which can result in the degradation of image quality. This paper proposes a method that leverages the unique optical characteristics of H&E stained images. A three-branch U-Net architecture has been implemented, where each branch contributes to the final segmentation results. The process includes applying watershed algorithm to separate overlapping regions and enhance accuracy. The Triple U-Net architecture comprises an RGB branch, a Hematoxylin branch, and a Segmentation branch. This study focuses on a novel dataset named CryoNuSeg. The results obtained through robust experiments outperform the state-of-the-art results across various metrics. The benchmark score for this dataset is AJI 52.5 and PQ 47.7, achieved through the implementation of U-Net Architecture. However, the proposed Triple U-Net architecture achieves an AJI score of 67.41 and PQ of 50.56. The proposed architecture improves more on AJI than other evaluation metrics, which further justifies the superiority of the Triple U-Net architecture over the baseline U-Net model, as AJI is a more strict evaluation metric. The use of the three-branch U-Net model, followed by watershed post-processing, significantly surpasses the benchmark scores, showing substantial improvement in the AJI score, Comment: To be published in "6th IVPR & 11th ICIEV"

Active matter systems being in a non-equilibrium state, exhibit complex behaviors such as self-organization and giving rise to emergent phenomena. There are many examples of active particles with biological origins, including bacteria and spermatozoa, or with artificial origins, such as self-propelled swimmers and Janus particles. The ability to manipulate active particles is vital for their effective application e.g. separating motile spermatozoa from nonmotile and dead ones, to increase fertilization chance. In this study, we proposed a mechanism -- an apparatus -- to sort and demix active particles based on their motility values (P\`eclet number). Initially, using Brownian simulations we demonstrated the feasibility of sorting self-propelled particles. Following this, we employed machine learning methods, supplemented with data from comprehensive simulations that we conducted for this study, to model the complex behavior of active particles. This enabled us to sort them based on their P\`eclet number. Finally, we evaluated the performance of the developed models and showed their effectiveness in demixing and sorting the active particles. Our findings can find applications in various fields, including physics, biology, and biomedical science, where the sorting and manipulation of active particles play a pivotal role., Comment: 13 pages, 5 figures

Today, our lifestyle revolves around digital devices powered by microprocessors of different instruction set architectures (ISA). Among them, the most common are x86 and ARM, the brainpower of our computers and smartphones. Reduced instruction set computing (RISC) is the basis of ARM architecture, designed to offer greater energy efficiency. On the other hand, principles of complex instruction set computing (CISC) are utilized by x86 processors, which handle heavier computing tasks while being more power-hungry. The rise of smartphones over a decade has changed the laptop market. It also influenced customers towards ARM-based energy-efficient laptops. This transition in the computing segment is seen not only in consumers but also in commercial settings, especially data centers. Usually, data centers are designed to operate 24/7, where energy is a big concern. So, ARM chips have started making their way to cloud servers. This paper comprehensively analyzes and compares the ARM and x86 architectures to unravel the factors contributing to ARM's increasing dominance in the market. It also explores the impact of smartphones on the laptop market and assesses the significance of system-on-a-chip (SoC). Focusing on the proper utilization of energy and sustainability, our paper offers valuable insights into the growing trend of adopting ARM processors in the computing industry., Comment: 19 pages, 13 figures

Generative Flow Networks (GFlowNets) are a family of probabilistic generative models that learn to sample compositional objects proportional to their rewards. One big challenge of GFlowNets is training them effectively when dealing with long time horizons and sparse rewards. To address this, we propose Evolution guided generative flow networks (EGFN), a simple but powerful augmentation to the GFlowNets training using Evolutionary algorithms (EA). Our method can work on top of any GFlowNets training objective, by training a set of agent parameters using EA, storing the resulting trajectories in the prioritized replay buffer, and training the GFlowNets agent using the stored trajectories. We present a thorough investigation over a wide range of toy and real-world benchmark tasks showing the effectiveness of our method in handling long trajectories and sparse rewards., Comment: 16 pages, 16 figues

Recent advances in Deep Learning and Computer Vision have been successfully leveraged to serve marginalized communities in various contexts. One such area is Sign Language - a primary means of communication for the deaf community. However, so far, the bulk of research efforts and investments have gone into American Sign Language, and research activity into low-resource sign languages - especially Bangla Sign Language - has lagged significantly. In this research paper, we present a new word-level Bangla Sign Language dataset - BdSL40 - consisting of 611 videos over 40 words, along with two different approaches: one with a 3D Convolutional Neural Network model and another with a novel Graph Neural Network approach for the classification of BdSL40 dataset. This is the first study on word-level BdSL recognition, and the dataset was transcribed from Indian Sign Language (ISL) using the Bangla Sign Language Dictionary (1997). The proposed GNN model achieved an F1 score of 89%. The study highlights the significant lexical and semantic similarity between BdSL, West Bengal Sign Language, and ISL, and the lack of word-level datasets for BdSL in the literature. We release the dataset and source code to stimulate further research.

Autonomous vehicles are one of the most popular and also fast-growing technologies in the world. As we go further, there are still a lot of challenges that are unsolved and may cause problems in the future when it comes to testing in real world. Simulations on the other hand have always had a huge impact in the fields of science, technology, physics, etc. The simulation also powers real-world Autonomous Vehicles nowadays. Therefore, We have built an Autonomous Vehicle Simulation Software - called AVIS Engine - that provides tools and features that help develop autonomous vehicles in various environments. AVIS Engine features an advanced input and output system for the vehicle and includes a traffic system and vehicle sensor system which can be communicated using the fast networking system and ROS Bridge., Comment: 13 pages, 7 figures

We present the discovery of twelve thus far non-repeating fast radio burst (FRB) sources, detected by the Canadian Hydrogen Intensity Mapping Experiment (CHIME) telescope. These sources were selected from a database comprising of order $10^3$ CHIME/FRB full-array raw voltage data recordings, based on their exceptionally high brightness and complex morphology. Our study examines the time-frequency characteristics of these bursts, including drifting, microstructure, and periodicities. The events in this sample display a variety of unique drifting phenomenologies that deviate from the linear negative drifting phenomenon seen in many repeating FRBs, and motivate a possible new framework for classifying drifting archetypes. Additionally, we detect microstructure features of duration $\lesssim$ 50 $\mu s$ in seven events, with some as narrow as $\approx$ 7 $\mu s$. We find no evidence of significant periodicities. Furthermore, we report the polarization characteristics of seven events, including their polarization fractions and Faraday rotation measures (RMs). The observed $|\mathrm{RM}|$ values span a wide range of $17.24(2)$ - $328.06(2) \mathrm{~rad~m}^{-2}$, with linear polarization fractions between $0.340(1)$ - $0.946(3)$. The morphological properties of the bursts in our sample appear broadly consistent with predictions from both relativistic shock and magnetospheric models of FRB emission, as well as propagation through discrete ionized plasma structures. We address these models and discuss how they can be tested using our improved understanding of morphological archetypes., Comment: 29 pages, 14 figures, submitted to ApJ, comments appreciated

Parkinson's disease (PD) is a neuro-degenerative disorder that affects movement, speech, and coordination. Timely diagnosis and treatment can improve the quality of life for PD patients. However, access to clinical diagnosis is limited in low and middle income countries (LMICs). Therefore, development of automated screening tools for PD can have a huge social impact, particularly in the public health sector. In this paper, we present PULSAR, a novel method to screen for PD from webcam-recorded videos of the finger-tapping task from the Movement Disorder Society - Unified Parkinson's Disease Rating Scale (MDS-UPDRS). PULSAR is trained and evaluated on data collected from 382 participants (183 self-reported as PD patients). We used an adaptive graph convolutional neural network to dynamically learn the spatio temporal graph edges specific to the finger-tapping task. We enhanced this idea with a multi stream adaptive convolution model to learn features from different modalities of data critical to detect PD, such as relative location of the finger joints, velocity and acceleration of tapping. As the labels of the videos are self-reported, there could be cases of undiagnosed PD in the non-PD labeled samples. We leveraged the idea of Positive Unlabeled (PU) Learning that does not need labeled negative data. Our experiments show clear benefit of modeling the problem in this way. PULSAR achieved 80.95% accuracy in validation set and a mean accuracy of 71.29% (2.49% standard deviation) in independent test, despite being trained with limited amount of data. This is specially promising as labeled data is scarce in health care sector. We hope PULSAR will make PD screening more accessible to everyone. The proposed techniques could be extended for assessment of other movement disorders, such as ataxia, and Huntington's disease.

In 2021, a catalog of 536 fast radio bursts (FRBs) detected with the Canadian Hydrogen Intensity Mapping Experiment (CHIME) radio telescope was released by the CHIME/FRB Collaboration. This large collection of bursts, observed with a single instrument and uniform selection effects, has advanced our understanding of the FRB population. Here we update the results for 140 of these FRBs for which channelized raw voltage ('baseband') data are available. With the voltages measured by the telescope's antennas, it is possible to maximize the telescope sensitivity in any direction within the primary beam, an operation called 'beamforming'. This allows us to increase the signal-to-noise ratio (S/N) of the bursts and to localize them to sub-arcminute precision. The improved localization is also used to correct the beam response of the instrument and to measure fluxes and fluences with a ~10% uncertainty. Additionally, the time resolution is increased by three orders of magnitude relative to that in the first CHIME/FRB catalog, and, applying coherent dedispersion, burst morphologies can be studied in detail. Polarization information is also available for the full sample of 140 FRBs, providing an unprecedented dataset to study the polarization properties of the population. We release the baseband data beamformed to the most probable position of each FRB. These data are analyzed in detail in a series of accompanying papers.

Due to limited resources and fast economic growth, designing optimal transportation road networks with traffic simulation and validation in a cost-effective manner is vital for developing countries, where extensive manual testing is expensive and often infeasible. Current rule-based road design generators lack diversity, a key feature for design robustness. Generative Flow Networks (GFlowNets) learn stochastic policies to sample from an unnormalized reward distribution, thus generating high-quality solutions while preserving their diversity. In this work, we formulate the problem of linking incident roads to the circular junction of a roundabout by a Markov decision process, and we leverage GFlowNets as the Junction-Art road generator. We compare our method with related methods and our empirical results show that our method achieves better diversity while preserving a high validity score., Comment: 6 pages. Submitted to ReALML@NeurIPS (2023)

Convolutional Neural Networks (CNNs) are used in a wide range of applications, with full-precision CNNs achieving high accuracy at the expense of portability. Recent progress in quantization techniques has demonstrated that sub-byte Quantized Neural Networks (QNNs) achieve comparable or superior accuracy while significantly reducing the computational cost and memory footprint. However, sub-byte computation on commodity hardware is sub-optimal due to the lack of support for such precision. In this paper, we introduce Sparq, a Sub-byte vector Processor designed for the AcceleRation of QNN inference. This processor is based on a modified version of Ara, an open-source 64-bit RISC-V ``V'' compliant processor. Sparq is implemented in GLOBAL FOUNDRIES 22FDX FD-SOI technology and extends the Instruction Set Architecture (ISA) by adding a new multiply-shift-accumulate instruction to improve sub-byte computation effciency. The floating-point unit is also removed to minimize area and power usage. To demonstrate Sparq performance, we implement an ultra-low-precision (1-bit to 4-bit) vectorized conv2d operation taking advantage of the dedicated hardware. We show that Sparq can significantly accelerate sub-byte computations with respectively 3.2 times, and 1.7 times acceleration over an optimized 16-bit 2D convolution for 2-bit and 4-bit quantization., Comment: 5 pages, Accepted for publication in the 21st IEEE Interregional NEWCAS Conference (2023)

Metastability in soft condensed matter systems usually results from the presence of a nucleation free energy barrier and/or slow dynamics caused by high density jamming phenomena. Here, we use molecular dynamics and Monte Carlo simulation to show that the interactions between topological defects stabilizes a chiral helical fluid in a confined quasi-one-dimensional hard sphere fluid dramatically slowing its decay toward the equilibrium achiral fluid state. Analysis of thermodynamic, structural and dynamic properties of the system show the equation of state bifurcates continuously at intermediate pressures into two distinct branches that are accessed from different initial conditions, but terminate at the same close packed single helix in the high pressure limit. The equilibrium fluid, which forms the high pressure branch as the system is compressed from low density, is characterized by helical sections separated by randomly distributed topological defects that change the helical twist direction giving rise to an achiral fluid. The low pressure metastable branch, formed by decompressing the system from the perfect helix, is characterized by the appearance of loosely paired defects that help retain the chiral excess of the original state and stabliize the fluids until it merges continuously with the equilibrium branch at intermediate pressures., Comment: 8 Figures, 2 figures in supplemental information

Today text classification becomes critical task for concerned individuals for numerous purposes. Hence, several researches have been conducted to develop automatic text classification for national and international languages. However, the need for an automatic text categorization system for local languages is felt. The main aim of this study is to establish a Pashto automatic text classification system. In order to pursue this work, we built a Pashto corpus which is a collection of Pashto documents due to the unavailability of public datasets of Pashto text documents. Besides, this study compares several models containing both statistical and neural network machine learning techniques including Multilayer Perceptron (MLP), Support Vector Machine (SVM), K Nearest Neighbor (KNN), decision tree, gaussian na\"ive Bayes, multinomial na\"ive Bayes, random forest, and logistic regression to discover the most effective approach. Moreover, this investigation evaluates two different feature extraction methods including unigram, and Time Frequency Inverse Document Frequency (IFIDF). Subsequently, this research obtained average testing accuracy rate 94% using MLP classification algorithm and TFIDF feature extraction method in this context., Comment: arXiv admin note: substantial text overlap with arXiv:2305.03201

Federated Learning (FL), a distributed machine learning technique has recently experienced tremendous growth in popularity due to its emphasis on user data privacy. However, the distributed computations of FL can result in constrained communication and drawn-out learning processes, necessitating the client-server communication cost optimization. The ratio of chosen clients and the quantity of local training passes are two hyperparameters that have a significant impact on FL performance. Due to different training preferences across various applications, it can be difficult for FL practitioners to manually select such hyperparameters. In our research paper, we introduce FedAVO, a novel FL algorithm that enhances communication effectiveness by selecting the best hyperparameters leveraging the African Vulture Optimizer (AVO). Our research demonstrates that the communication costs associated with FL operations can be substantially reduced by adopting AVO for FL hyperparameter adjustment. Through extensive evaluations of FedAVO on benchmark datasets, we show that FedAVO achieves significant improvement in terms of model accuracy and communication round, particularly with realistic cases of Non-IID datasets. Our extensive evaluation of the FedAVO algorithm identifies the optimal hyperparameters that are appropriately fitted for the benchmark datasets, eventually increasing global model accuracy by 6% in comparison to the state-of-the-art FL algorithms (such as FedAvg, FedProx, FedPSO, etc.)., Comment: 8 pages

High-definition roads are an essential component of realistic driving scenario simulation for autonomous vehicle testing. Roundabouts are one of the key road segments that have not been thoroughly investigated. Based on the geometric constraints of the nearby road structure, this work presents a novel method for procedurally building roundabouts. The suggested method can result in roundabout lanes that are not perfectly circular and resemble real-world roundabouts by allowing approaching roadways to be connected to a roundabout at any angle. One can easily incorporate the roundabout in their HD road generation process or use the standalone roundabouts in scenario-based testing of autonomous driving., Comment: (6 Pages) Accepted at IEEE Intelligent Vehicles Symposium 2023

Named Entity Recognition (NER) is a fundamental task in natural language processing that involves identifying and classifying named entities in text. But much work hasn't been done for complex named entity recognition in Bangla, despite being the seventh most spoken language globally. CNER is a more challenging task than traditional NER as it involves identifying and classifying complex and compound entities, which are not common in Bangla language. In this paper, we present the winning solution of Bangla Complex Named Entity Recognition Challenge - addressing the CNER task on BanglaCoNER dataset using two different approaches, namely Conditional Random Fields (CRF) and finetuning transformer based Deep Learning models such as BanglaBERT. The dataset consisted of 15300 sentences for training and 800 sentences for validation, in the .conll format. Exploratory Data Analysis (EDA) on the dataset revealed that the dataset had 7 different NER tags, with notable presence of English words, suggesting that the dataset is synthetic and likely a product of translation. We experimented with a variety of feature combinations including Part of Speech (POS) tags, word suffixes, Gazetteers, and cluster information from embeddings, while also finetuning the BanglaBERT (large) model for NER. We found that not all linguistic patterns are immediately apparent or even intuitive to humans, which is why Deep Learning based models has proved to be the more effective model in NLP, including CNER task. Our fine tuned BanglaBERT (large) model achieves an F1 Score of 0.79 on the validation set. Overall, our study highlights the importance of Bangla Complex Named Entity Recognition, particularly in the context of synthetic datasets. Our findings also demonstrate the efficacy of Deep Learning models such as BanglaBERT for NER in Bangla language., Comment: Winning Solution for the Bangla Complex Named Entity Recognition Challenge