Md Kowsher, Tara Esmaeilbeig, Chun-Nam Yu, Chen Chen, Mojtaba Soltanalian, Niloofar Yousefi. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Md Kowsher, Nusrat Jahan Prottasha, Prakash Bhat, Chun-Nam Yu, Mojtaba Soltanalian, Ivan Garibay, Ozlem Garibay, Chen Chen, Niloofar Yousefi. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

LTE evolved Multimedia Broadcast/Multicast Service (eMBMS) is an attractive solution for video delivery to very large groups in crowded venues. However, the deployment and management of eMBMS systems are challenging, due to the lack of real-time feedback from the user equipments (TIEs). Therefore, we present the Dynamic Monitoring (DyMo) system for low-overhead feedback collection. DyMo leverages eMBMS for broadcasting stochastic group instructions to all UEs. These instructions indicate the reporting rates as a function of the observed quality of service (QoS). This simple feedback mechanism collects very limited QoS reports from the TIEs. The reports are used for network optimization, thereby ensuring high QoS to the TIEs. We present the design aspects of DyMo and evaluate its performance analytically and via extensive simulations. Specifically, we show that DyMo infers the optimal eMBMS settings with extremely low overhead while meeting strict QoS requirements under different TIE mobility patterns and presence of network component failures. For instance, DyMo can detect the eMBMS signal-to-noise ratio experienced by the 0.1th percentile of the TIEs with a root mean square error of 0.05% with only 5 to 10 reports per second regardless of the number of TIEs.

We present a method of extracting figures and images from the pages of scanned documents, especially from technical research articles. Our approach is novel in two key ways. First, we treat this as a computer vision problem, and train convolutional neural networks to recognize figures in scanned pages. Second, we generate our training data from 'born-digital' structured documents, allowing us to automatically produce labels for our training set using PDF figure extractors. This avoids the otherwise tedious task of hand-labelling thousands of document pages. Our convolutional neural networks achieve precision and recall of close to 85% in identifying figures from a test set consisting of modern journal papers and conference proceedings, and obtain precision and recall above 80% on an application data set comprised of historical technical documents scanned from the Bell Labs Records. Our results show that models trained on digital documents transfer very well to historical scans. Finally, it is easy to extend our models to identify other document elements such as tables and captions.

LTE evolved Multimedia Broadcast/Multicast Service (eMBMS) is an attractive solution for video delivery to very large groups in crowded venues. However, deployment and management of eMBMS systems is challenging, due to the lack of realtime feedback from the User Equipment (UEs). Therefore, we present the Dynamic Monitoring (DyMo) system for low-overhead feedback collection. DyMo leverages eMBMS for broadcasting Stochastic Group Instructions to all UEs. These instructions indicate the reporting rates as a function of the observed Quality of Service (QoS). This simple feedback mechanism collects very limited QoS reports from the UEs. The reports are used for network optimization, thereby ensuring high QoS to the UEs. We present the design aspects of DyMo and evaluate its performance analytically and via extensive simulations. Specifically, we show that DyMo infers the optimal eMBMS settings with extremely low overhead, while meeting strict QoS requirements under different UE mobility patterns and presence of network component failures. For instance, DyMo can detect the eMBMS Signal-to-Noise Ratio (SNR) experienced by the 0.1% percentile of the UEs with Root Mean Square Error (RMSE) of 0.05% with only 5 to 10 reports per second regardless of the number of UEs.

Many learning situations involve learning the conditional distribution p(y|x) when the training instances are drawn from the training distribution p_tr(x), even though it will later be used to predict for instances drawn from a different test distribution p_te(x). Most current approaches focus on learning how to reweigh the training examples, to make them resemble the test distribution. However, reweighing does not always help, because (we show that) the test error also depends on the correctness of the underlying model class. This paper analyses this situation by viewing the problem of learning under changing distributions as a game between a learner and an adversary. We characterize when such reweighing is needed, and also provide an algorithm, robust covariate shift adjustment (RCSA), that provides relevant weights. Our empirical studies, on UCI datasets and a real-world cancer prognostic prediction dataset, show that our analysis applies, and that our RCSA works effectively.

An accurate model of patient survival time can help in the treatment and care of cancer patients. The common practice of providing survival time estimates based only on population averages for the site and stage of cancer ignores many important individual differences among patients. In this paper, we propose a local regression method for learning patient-specific survival time distribution based on patient attributes such as blood tests and clinical assessments. When tested on a cohort of more than 2000 cancer patients, our method gives survival time predictions that are much more accurate than popular survival analysis models such as the Cox and Aalen regression models. Our results also show that using patient-specific attributes can reduce the prediction error on survival time by as much as 20% when compared to using cancer site and stage only.