Paper Accepted in PACLIC 37

/in /by

We are happy to announce that one of our papers titled “EDAL: Entropy based Dynamic Attention Loss for HateSpeech Classification”, has been accepted in the “37th Pacific Asia Conference on Language, Information and Computation”. 

In this work, the authors introduce the “Entropy-based Dynamic Attention Loss” (EDAL) to enhance model interpretability by incorporating an additional attention layer. EDAL encourages attention scores that provide valuable insights and boosts the performance of pretrained models during fine-tuning for downstream tasks. We conduct extensive experiments on six diverse datasets, confirming that EDAL effectively enhances classification performance while maintaining interpretability. Additionally, experiments with various pretrained models demonstrate EDAL’s significant performance improvements during fine-tuning. In summary, EDAL holds promise for creating more transparent and reliable hate speech classifiers, contributing to a safer online environment.

RA Recruitment Announcement 📣

/in /by
Read more

CCDS RA ranked #1 in BanglaNLP workshop @EMNLP23

/in /by

CCDS RA, Md Fahim has ranked #1 and #2 in Bangla sentiment analysis and violence inciting text detection tasks respectively organized by the first Bangla Language Processing workshop @ EMNLP 2023. More information about the workshop can be found here: First Workshop on Bangla Language Processing (blp-workshop.github.io) 

Congratulations to Fahim.

CCDS RA receives fully funded MS admission in the Concordia University

/in /by

Congratulations to Mohammad Raihanul Bashar Fahim! We’re happy to announce that he will be joining the Extended Reality and Interaction Technologies (EXiT) Lab at Concordia University, Canada for his Master’s in Computer Science starting in Fall 2023. He will be working in the field of Extended Reality, Human-Computer Interaction, and 3D user interface. Fahim had been one of our first RAs and an alumni of CSE IUB and has been working as a senior ML engineer since he left the lab.

Best of luck, Fahim!

CCDS RA receives fully funded PhD admission in the University of Houston

/in /by

Congratulations to Ovi Paul, one of our former RAs at CCDS and alumni of IUB, for embarking on his PhD journey. He will be doing his PhD in Geosensing System Engineering & Science program at the University of Houston! Ovi will also serve as a Graduate Research Assistant at the National Center for Airborne Laser Mapping. The center is based at the University of Houston and is operated in partnership with the University of California, Berkeley. At CCDS, he had been working on spatial data processing (land cover and land usage classification, urban buildup classification etc. from satellite data).

We wish him all the best.

Accepted Paper: Radio Galaxy Classification at INNS DLIA 2023

/in , /by

Our research paper, ‘Morphological Classification of Radio Galaxies using Semi-Supervised Group Equivariant CNNs,’ has been accepted for presentation at the esteemed INNS Deep Learning Innovations and Applications (INNS DLIA 2023) workshop, which is part of the International Joint Conference on Neural Networks (IJCNN 2023). The paper will also be published in the renowned Procedia Computer Science journal!

In this study, we tackled the challenge of limited labeled data in radio galaxy classification by employing a cutting-edge semi-supervised learning approach. By harnessing the power of Group Equivariant Convolutional Neural Networks (G-CNNs) as encoders, we achieved impressive results in classifying radio galaxies into the well-known Fanaroff-Riley Type I (FRI) and Type II (FRII) categories. [Link to Paper]

Explainable Hate Speech Detection: ICML 2023 Workshop Acceptance

/in , /by

Recently, Md Fahim, RA of CCDS with co-authors from UToronto, IUT and Fordham University has a paper accepted in AI and HCI workshop of ICML 2023. The paper proposes an interpretability and explainability oriented model to detect hate speech utilizing the pre-trained large language models. It creates dynamic class specific conceptual subspaces from which class specific attention is obtained by projecting the contextual embedding onto those spaces. These attentions provide better explainability of the detection task.
Paper Link: HateXplain2.0: An Explainable Hate Speech Detection Framework Utilizing Subjective Projection from Contextual Knowledge Space to Disjoint Concept Space

CCDS RA Selected for the MIT Summer Geometry Initiative

/in /by

One of our RAs Munshi Sanowar Raihan has been selected for the MIT Summer Geometry Initiative (a six week research program) introducing undergrad and graduate students to the field of geometric processing. He is among the 32 fellows selected from all over the world. The program, organized by Prof Justine Solomon, PI, MIT’s geometry processing groups, provides students with the opportunity to collaborate with program staff that include graduate students, faculty and research scientists.

CCDS RA receives full funded MS admission in the Illinois Institute of Technology, Illinois, USA

/in /by

Congrats to Samin Bin Karim, now RA of CCDS and alumni of CSE, IUB, for starting his graduate studies in Illinois Institute of Technology, Illinois, USA. We wish him the very best of luck. Samin had been working on reinforcement learning and multi-agent systems here at CCDS.