Prasenjit Mitra

Nan Zhang, Yusen Zhang, Wu Guo et al.

Summaries of medical text shall be faithful by being consistent and factual with source inputs, which is an important but understudied topic for safety and efficiency in healthcare. In this paper, we investigate and improve faithfulness in summarization on a broad range of medical summarization tasks. Our investigation reveals that current summarization models often produce unfaithful outputs for medical input text. We then introduce FaMeSumm, a framework to improve faithfulness by fine-tuning pre-trained language models based on medical knowledge. FaMeSumm performs contrastive learning on designed sets of faithful and unfaithful summaries, and it incorporates medical terms and their contexts to encourage faithful generation of medical terms. We conduct comprehensive experiments on three datasets in two languages: health question and radiology report summarization datasets in English, and a patient-doctor dialogue dataset in Chinese. Results demonstrate that FaMeSumm is flexible and effective by delivering consistent improvements over mainstream language models such as BART, T5, mT5, and PEGASUS, yielding state-of-the-art performances on metrics for faithfulness and general quality. Human evaluation by doctors also shows that FaMeSumm generates more faithful outputs. Our code is available at https://github.com/psunlpgroup/FaMeSumm .

Victor Tolulope Olufemi, Oreoluwa Babatunde, Ramsey Njema et al.

We evaluate whether compact domain-specialized ASR models can outperform massively multilingual foundation models for conversational African speech across 19 languages in the WAXAL corpus. Fine-tuned edge models achieve a macro-averaged WER of $38.0\%$ compared to $64.9\%$ for the best zero-shot baseline, a $26.9$ percentage-point reduction using models $3-40\times$ smaller. Results confirm that domain specialization dominates scale for spontaneous African speech. Cross-domain evaluation shows that fine-tuned models recover usable performance on out-of-distribution (OOD) speech, while zero-shot models regain an advantage when the test domain matches their pretraining distribution. A distributed native-speaker audit across all surveyed languages produces a linguistically-grounded error taxonomy, showing that CTC and autoregressive architectures behave differently across language families. We further show that WER alone misrepresents performance for syllabary-script languages where CER/WER ratios reveal substantially higher character-level accuracy than headline WER suggests. Finally, to contribute to future African ASR research, we release all model weights, fine-tuning and evaluation scripts, and a cleaned WAXAL subset covering all $19$ languages.

Scott Pezanowski, Prasenjit Mitra, Alan M. MacEachren

Sensemaking using automatically extracted information from text is a challenging problem. In this paper, we address a specific type of information extraction, namely extracting information related to descriptions of movement. Aggregating and understanding information related to descriptions of movement and lack of movement specified in text can lead to an improved understanding and sensemaking of movement phenomena of various types, e.g., migration of people and animals, impediments to travel due to COVID-19, etc. We present GeoMovement, a system that is based on combining machine learning and rule-based extraction of movement-related information with state-of-the-art visualization techniques. Along with the depiction of movement, our tool can extract and present a lack of movement. Very little prior work exists on automatically extracting descriptions of movement, especially negation and movement. Apart from addressing these, GeoMovement also provides a novel integrated framework for combining these extraction modules with visualization. We include two systematic case studies of GeoMovement that show how humans can derive meaningful geographic movement information. GeoMovement can complement precise movement data, e.g., obtained using sensors, or be used by itself when precise data is unavailable.

Amogh Subbakrishna Adishesha, Lily Jakielaszek, Fariha Azhar et al.

The increasing reliance on online communities for healthcare information by patients and caregivers has led to the increase in the spread of misinformation, or subjective, anecdotal and inaccurate or non-specific recommendations, which, if acted on, could cause serious harm to the patients. Hence, there is an urgent need to connect users with accurate and tailored health information in a timely manner to prevent such harm. This paper proposes an innovative approach to suggesting reliable information to participants in online communities as they move through different stages in their disease or treatment. We hypothesize that patients with similar histories of disease progression or course of treatment would have similar information needs at comparable stages. Specifically, we pose the problem of predicting topic tags or keywords that describe the future information needs of users based on their profiles, traces of their online interactions within the community (past posts, replies) and the profiles and traces of online interactions of other users with similar profiles and similar traces of past interaction with the target users. The result is a variant of the collaborative information filtering or recommendation system tailored to the needs of users of online health communities. We report results of our experiments on an expert curated data set which demonstrate the superiority of the proposed approach over the state of the art baselines with respect to accurate and timely prediction of topic tags (and hence information sources of interest).

Shreya Ghosh, Saptarshi Sengupta, Prasenjit Mitra

In this paper, we lay out a vision for analysing semantic trajectory traces and generating synthetic semantic trajectory data (SSTs) using generative language model. Leveraging the advancements in deep learning, as evident by progress in the field of natural language processing (NLP), computer vision, etc. we intend to create intelligent models that can study the semantic trajectories in various contexts, predicting future trends, increasing machine understanding of the movement of animals, humans, goods, etc. enhancing human-computer interactions, and contributing to an array of applications ranging from urban-planning to personalized recommendation engines and business strategy.

Jakob Hederich, Shreya Ghosh, Zeyu He et al.

We introduce NightPulse, an interactive tool for Night-time light (NTL) data visualization and analytics, which enables researchers and stakeholders to explore and analyze NTL data with a user-friendly platform. Powered by efficient system architecture, NightPulse supports image segmentation, clustering, and change pattern detection to identify urban development and sprawl patterns. It captures temporal trends of NTL and semantics of cities, answering questions about demographic factors, city boundaries, and unusual differences.

Matthew Hines, Gregory Glatzer, Shreya Ghosh et al.

The interaction between elephants and their environment has profound implications for both ecology and conservation strategies. This study presents an analytical approach to decipher the intricate patterns of elephant movement in Sub-Saharan Africa, concentrating on key ecological drivers such as seasonal variations and rainfall patterns. Despite the complexities surrounding these influential factors, our analysis provides a holistic view of elephant migratory behavior in the context of the dynamic African landscape. Our comprehensive approach enables us to predict the potential impact of these ecological determinants on elephant migration, a critical step in establishing informed conservation strategies. This projection is particularly crucial given the impacts of global climate change on seasonal and rainfall patterns, which could substantially influence elephant movements in the future. The findings of our work aim to not only advance the understanding of movement ecology but also foster a sustainable coexistence of humans and elephants in Sub-Saharan Africa. By predicting potential elephant routes, our work can inform strategies to minimize human-elephant conflict, effectively manage land use, and enhance anti-poaching efforts. This research underscores the importance of integrating movement ecology and climatic variables for effective wildlife management and conservation planning.

Tianrong Zhang, Bochuan Cao, Yuanpu Cao et al.

The recent breakthrough in large language models (LLMs) such as ChatGPT has revolutionized production processes at an unprecedented pace. Alongside this progress also comes mounting concerns about LLMs' susceptibility to jailbreaking attacks, which leads to the generation of harmful or unsafe content. While safety alignment measures have been implemented in LLMs to mitigate existing jailbreak attempts and force them to become increasingly complicated, it is still far from perfect. In this paper, we analyze the common pattern of the current safety alignment and show that it is possible to exploit such patterns for jailbreaking attacks by simultaneous obfuscation in queries and responses. Specifically, we propose WordGame attack, which replaces malicious words with word games to break down the adversarial intent of a query and encourage benign content regarding the games to precede the anticipated harmful content in the response, creating a context that is hardly covered by any corpus used for safety alignment. Extensive experiments demonstrate that WordGame attack can break the guardrails of the current leading proprietary and open-source LLMs, including the latest Claude-3, GPT-4, and Llama-3 models. Further ablation studies on such simultaneous obfuscation in query and response provide evidence of the merits of the attack strategy beyond an individual attack.

Nan Zhang, Yanchi Liu, Xujiang Zhao et al.

Large Language Models (LLMs) have exhibited remarkable proficiency across a wide array of NLP tasks. However, the escalation in model size also engenders substantial deployment costs. While few efforts have explored model pruning techniques to reduce the size of LLMs, they mainly center on general or task-specific weights. This leads to suboptimal performance due to lacking specificity on the target domain or generality on different tasks when applied to domain-specific challenges. This work introduces an innovative unstructured dual-pruning methodology, D-Pruner, for domain-specific compression on LLM. It extracts a compressed, domain-specific, and task-agnostic LLM by identifying LLM weights that are pivotal for general capabilities, like linguistic capability and multi-task solving, and domain-specific knowledge. More specifically, we first assess general weight importance by quantifying the error incurred upon their removal with the help of an open-domain calibration dataset. Then, we utilize this general weight importance to refine the training loss, so that it preserves generality when fitting into a specific domain. Moreover, by efficiently approximating weight importance with the refined training loss on a domain-specific calibration dataset, we obtain a pruned model emphasizing generality and specificity. Our comprehensive experiments across various tasks in healthcare and legal domains show the effectiveness of D-Pruner in domain-specific compression. Our code is available at https://github.com/psunlpgroup/D-Pruner.

Nan Zhang, Prafulla Kumar Choubey, Alexander Fabbri et al.

Indexing is an important step towards strong performance in retrieval-augmented generation (RAG) systems. However, existing methods organize data based on either semantic similarity (similarity) or related information (relatedness), but do not cover both perspectives comprehensively. Our analysis reveals that modeling only one perspective results in insufficient knowledge synthesis, leading to suboptimal performance on complex tasks requiring multihop reasoning. In this paper, we propose SiReRAG, a novel RAG indexing approach that explicitly considers both similar and related information. On the similarity side, we follow existing work and explore some variances to construct a similarity tree based on recursive summarization. On the relatedness side, SiReRAG extracts propositions and entities from texts, groups propositions via shared entities, and generates recursive summaries to construct a relatedness tree. We index and flatten both similarity and relatedness trees into a unified retrieval pool. Our experiments demonstrate that SiReRAG consistently outperforms state-of-the-art indexing methods on three multihop datasets (MuSiQue, 2WikiMultiHopQA, and HotpotQA), with an average 1.9% improvement in F1 scores. As a reasonably efficient solution, SiReRAG enhances existing reranking methods significantly, with up to 7.8% improvement in average F1 scores. Our code is available at https://github.com/SalesforceAIResearch/SiReRAG .

Mahounan Pericles Adjovi, Roald Eiselen, Prasenjit Mitra

Data scarcity limits NLP development for low-resource African languages. We evaluate two data augmentation methods -- LLM-based generation (Gemini 2.5 Flash) and back-translation (NLLB-200) -- for Hausa and Fongbe, two West African languages that differ substantially in LLM generation quality. We assess augmentation on named entity recognition (NER) and part-of-speech (POS) tagging using MasakhaNER 2.0 and MasakhaPOS benchmarks. Our results reveal that augmentation effectiveness depends on task type rather than language or LLM quality alone. For NER, neither method improves over baseline for either language; LLM augmentation reduces Hausa NER by 0.24% F1 and Fongbe NER by 1.81% F1. For POS tagging, LLM augmentation improves Fongbe by 0.33% accuracy, while back-translation improves Hausa by 0.17%; back-translation reduces Fongbe POS by 0.35% and has negligible effect on Hausa POS. The same LLM-generated synthetic data produces opposite effects across tasks for Fongbe -- hurting NER while helping POS -- suggesting task structure governs augmentation outcomes more than synthetic data quality. These findings challenge the assumption that LLM generation quality predicts augmentation success, and provide actionable guidance: data augmentation should be treated as a task-specific intervention rather than a universally beneficial preprocessing step.

Mahounan Pericles Adjovi, Roald Eiselen, Prasenjit Mitra

Large language models (LLMs) are trained on data contributed by low-resource language communities, yet the linguistic knowledge encoded in these models remains accessible only through commercial APIs. This paper investigates whether strategic prompting can extract usable text data from LLMs for two West African languages: Hausa (Afroasiatic, approximately 80 million speakers) and Fongbe (Niger-Congo, approximately 2 million speakers). We systematically compare six elicitation task types across two commercial LLMs (GPT-4o Mini and Gemini 2.5 Flash). GPT-4o Mini extracts 6-41 times more usable target-language words per API call than Gemini. Optimal strategies differ by language: Hausa benefits from functional text and dialogue, while Fongbe requires constrained generation prompts. We release all generated corpora and code.

Saptarshi Sengupta, Connor Heaton, Suhan Cui et al.

In Natural Language Processing (NLP), Machine Reading Comprehension (MRC) is the task of answering a question based on a given context. To handle questions in the medical domain, modern language models such as BioBERT, SciBERT and even ChatGPT are trained on vast amounts of in-domain medical corpora. However, in-domain pre-training is expensive in terms of time and resources. In this paper, we propose a resource-efficient approach for injecting domain knowledge into a model without relying on such domain-specific pre-training. Knowledge graphs are powerful resources for accessing medical information. Building on existing work, we introduce a method using Multi-Layer Perceptrons (MLPs) for aligning and integrating embeddings extracted from medical knowledge graphs with the embedding spaces of pre-trained language models (LMs). The aligned embeddings are fused with open-domain LMs BERT and RoBERTa that are fine-tuned for two MRC tasks, span detection (COVID-QA) and multiple-choice questions (PubMedQA). We compare our method to prior techniques that rely on a vocabulary overlap for embedding alignment and show how our method circumvents this requirement to deliver better performance. On both datasets, our method allows BERT/RoBERTa to either perform on par (occasionally exceeding) with stronger domain-specific models or show improvements in general over prior techniques. With the proposed approach, we signal an alternative method to in-domain pre-training to achieve domain proficiency. Our code is available here.

Suhan Cui, Prasenjit Mitra

In the realm of big data and digital healthcare, Electronic Health Records (EHR) have become a rich source of information with the potential to improve patient care and medical research. In recent years, machine learning models have proliferated for analyzing EHR data to predict patients future health conditions. Among them, some studies advocate for multi-task learning (MTL) to jointly predict multiple target diseases for improving the prediction performance over single task learning. Nevertheless, current MTL frameworks for EHR data have significant limitations due to their heavy reliance on human experts to identify task groups for joint training and design model architectures. To reduce human intervention and improve the framework design, we propose an automated approach named AutoDP, which can search for the optimal configuration of task grouping and architectures simultaneously. To tackle the vast joint search space encompassing task combinations and architectures, we employ surrogate model-based optimization, enabling us to efficiently discover the optimal solution. Experimental results on real-world EHR data demonstrate the efficacy of the proposed AutoDP framework. It achieves significant performance improvements over both hand-crafted and automated state-of-the-art methods, also maintains a feasible search cost at the same time. Source code can be found via the link: \url{https://github.com/SH-Src/AutoDP}.

Tewodros Kederalah Idris, Roald Eiselen, Prasenjit Mitra

Cross-lingual transfer learning enables NLP for low-resource languages by leveraging labeled data from higher-resource sources, yet existing comparisons of source language selection strategies do not control for total training data, confounding language selection effects with data quantity effects. We introduce Budget-Xfer, a framework that formulates multi-source cross-lingual transfer as a budget-constrained resource allocation problem. Given a fixed annotation budget B, our framework jointly optimizes which source languages to include and how much data to allocate from each. We evaluate four allocation strategies across named entity recognition and sentiment analysis for three African target languages (Hausa, Yoruba, Swahili) using two multilingual models, conducting 288 experiments. Our results show that (1) multi-source transfer significantly outperforms single-source transfer (Cohen's d = 0.80 to 1.98), driven by a structural budget underutilization bottleneck; (2) among multi-source strategies, differences are modest and non-significant; and (3) the value of embedding similarity as a selection proxy is task-dependent, with random selection outperforming similarity-based selection for NER but not sentiment analysis.

Blessed Madukoma, Prasenjit Mitra

Anti-doping programs rely on biological testing to detect performance-enhancing drugs, but such testing costs over $800 per sample and is limited by short detection windows for many prohibited substances. These constraints leave large portions of athletes without regular testing, motivating complementary screening approaches that analyze routine competition results to identify suspicious performance patterns. We present a system that processes 1.6 million athletics performances from over 19,000 competitions (2010-2025) using eight detection methods ranging from statistical rules to machine learning and trajectory analysis. We validate all methods against publicly confirmed anti-doping violations to measure their effectiveness in identifying sanctioned athletes. Trajectory-based methods, which compare performances to expected career progression, achieve the best balance between detecting violations and limiting false alarms, though all methods face challenges from incomplete data and rare confirmed violations. The system provides an interactive interface for expert-driven investigation, emphasizing transparency and human judgment to support, rather than replace, established anti-doping processes.

Edward Ajayi, Prasenjit Mitra

Humor generation poses a significant challenge for Large Language Models (LLMs), because their standard training objective - predicting the most likely next word - inherently conflicts with the surprise and incongruity needed for comedy. To bridge this gap, we introduce the Cognitive Synergy Framework, a theoretically grounded methodology for generating high-quality humor data inspired by psychological theories of humor. Utilizing a Mixture-of-Thought (MoT) approach, we deploy six cognitive personas (e.g., The Absurdist, The Cynic) to synthesize diverse comedic perspectives for a given prompt. This framework creates a theoretically grounded dataset, which we use to fine-tune a 7B-parameter student model. We compare Direct Preference Optimization (DPO) and a novel Offline Group Relative Policy Optimization (O-GRPO); our 7B model significantly outperforms larger instruction-tuned baselines and achieves performance competitive with state-of-the-art proprietary models. We find that cognitive-driven data curation is far more critical than alignment algorithms or model scale for humor generation. Code and data will be available upon publication.

Nan Zhang, Connor Heaton, Sean Timothy Okonsky et al.

Optical Character Recognition (OCR) is an established task with the objective of identifying the text present in an image. While many off-the-shelf OCR models exist, they are often trained for either scientific (e.g., formulae) or generic printed English text. Extracting text from chemistry publications requires an OCR model that is capable in both realms. Nougat, a recent tool, exhibits strong ability to parse academic documents, but is unable to parse tables in PubMed articles, which comprises a significant part of the academic community and is the focus of this work. To mitigate this gap, we present the Printed English and Chemical Equations (PEaCE) dataset, containing both synthetic and real-world records, and evaluate the efficacy of transformer-based OCR models when trained on this resource. Given that real-world records contain artifacts not present in synthetic records, we propose transformations that mimic such qualities. We perform a suite of experiments to explore the impact of patch size, multi-domain training, and our proposed transformations, ultimately finding that models with a small patch size trained on multiple domains using the proposed transformations yield the best performance. Our dataset and code is available at https://github.com/ZN1010/PEaCE.

Edward Ajayi, Prasenjit Mitra

Evaluating humor in large language models (LLMs) is an open challenge because existing approaches yield isolated, incomparable metrics rather than unified model rankings, making it difficult to track progress across systems. We introduce HumorRank, a tournament-based evaluation framework and leaderboard for textual humor generation. Using SemEval-2026 MWAHAHA test dataset, we conduct an extensive automated pairwise evaluation across nine models spanning proprietary, open-weight, and specialized systems. Pairwise judgments grounded in the General Theory of Verbal Humor (GTVH) are aggregated via an Adaptive Swiss tournament, with Bradley-Terry Maximum Likelihood Estimation (MLE) producing globally consistent humor generation capability rankings. Our results demonstrate that HumorRank yields statistically grounded model stratifications, showing that humor quality is driven by mastery of comedic mechanisms rather than model scale alone. HumorRank thus provides a scalable, interpretable methodology for benchmarking and understanding LLM-generated humor.

Ali Al-Lawati, Jason Lucas, Prasenjit Mitra

Large Language Models (LLMs) have demonstrated remarkable performance in various NLP tasks, including semantic parsing, which translates natural language into formal code representations. However, the reverse process, translating code into natural language, termed semantic captioning, has received less attention. This task is becoming increasingly important as LLMs are integrated into platforms for code generation, security analysis, and educational purposes. In this paper, we focus on the captioning of SQL query (SQL2Text) to address the critical need for understanding and explaining SQL queries in an era where LLM-generated code poses potential security risks. We repurpose Text2SQL datasets for SQL2Text by introducing an iterative ICL prompt using GPT-4o to generate multiple additional utterances, which enhances the robustness of the datasets for the reverse task. We conduct our experiments using in-context learning (ICL) based on different sample selection methods, emphasizing smaller, more computationally efficient LLMs. Our findings demonstrate that leveraging the inherent graph properties of SQL for ICL sample selection significantly outperforms random selection by up to 39% on BLEU score and provides better results than alternative methods. Dataset and codes are published: https://github.com/aliwister/ast-icl.

Ali Al-Lawati, Elsayed Eshra, Prasenjit Mitra

Trajectory generation is an important task in movement studies; it circumvents the privacy, ethical, and technical challenges of collecting real trajectories from the target population. In particular, real trajectories in the wildlife domain are scarce as a result of ethical and environmental constraints of the collection process. In this paper, we consider the problem of generating long-horizon trajectories, akin to wildlife migration, based on a small set of real samples. We propose a hierarchical approach to learn the global movement characteristics of the real dataset and recursively refine localized regions. Our solution, WildGraph, discretizes the geographic path into a prototype network of H3 (https://www.uber.com/blog/h3/) regions and leverages a recurrent variational auto-encoder to probabilistically generate paths over the regions, based on occupancy. WildGraph successfully generates realistic months-long trajectories using a sample size as small as 60. Experiments performed on two wildlife migration datasets demonstrate that our proposed method improves the generalization of the generated trajectories in comparison to existing work while achieving superior or comparable performance in several benchmark metrics. Our code is published on the following repository: https://github.com/aliwister/wildgraph.

Xianfeng Tang, Yandong Li, Yiwei Sun et al.

Graph neural networks (GNNs) are widely used in many applications. However, their robustness against adversarial attacks is criticized. Prior studies show that using unnoticeable modifications on graph topology or nodal features can significantly reduce the performances of GNNs. It is very challenging to design robust graph neural networks against poisoning attack and several efforts have been taken. Existing work aims at reducing the negative impact from adversarial edges only with the poisoned graph, which is sub-optimal since they fail to discriminate adversarial edges from normal ones. On the other hand, clean graphs from similar domains as the target poisoned graph are usually available in the real world. By perturbing these clean graphs, we create supervised knowledge to train the ability to detect adversarial edges so that the robustness of GNNs is elevated. However, such potential for clean graphs is neglected by existing work. To this end, we investigate a novel problem of improving the robustness of GNNs against poisoning attacks by exploring clean graphs. Specifically, we propose PA-GNN, which relies on a penalized aggregation mechanism that directly restrict the negative impact of adversarial edges by assigning them lower attention coefficients. To optimize PA-GNN for a poisoned graph, we design a meta-optimization algorithm that trains PA-GNN to penalize perturbations using clean graphs and their adversarial counterparts, and transfers such ability to improve the robustness of PA-GNN on the poisoned graph. Experimental results on four real-world datasets demonstrate the robustness of PA-GNN against poisoning attacks on graphs. Code and data are available here: https://github.com/tangxianfeng/PA-GNN.

Obed Irakoze, Prasenjit Mitra

Inferring the identity of wildlife species from daily movement data alone is a challenging task. We train sequence models on large-scale, 7-species GPS trajectories from the Movebank platform. Trajectories models are evaluated using a protocol in which entire telemetry studies or regions are heldout during testing. We compare Transformer-based sequence models to LSTM, CNN, and Temporal Convolutional Networks, and find that Transformers consistently achieve higher balanced accuracy with gains of approximately 8 to 22 percentage points, depending on the species and experimental setting. In an elephant binary classification task with 1-hour resolution, the Transformer achieves a balanced accuracy of 0.83 and an AUC of 0.92, substantially outperforming all baseline models. We examine, under data-limited conditions, feature representations by analyzing the differences between a basic displacement-based encoding and an expanded range of movement descriptors that include speed, direction, and turning behavior. With feature augmentation, we see clear performance gains, especially for underrepresented and sparsely represented species, such as large carnivores, lions, and Zebras. Finally, experiments comparing 1-hour and 30-minutetemporal resolutions show that while finer sampling can capture short-term movement patterns for some species, a unified 1-hour resolution yields more promising performance across studies by reducing missing data and ensuring consistent temporal coverage.

Ali Rasekh, Reza Heidari, Amir Hosein Haji Mohammad Rezaie et al.

With the increasing availability of diverse data types, particularly images and time series data from medical experiments, there is a growing demand for techniques designed to combine various modalities of data effectively. Our motivation comes from the important areas of predicting mortality and phenotyping where using different modalities of data could significantly improve our ability to predict. To tackle this challenge, we introduce a new method that uses two separate encoders, one for each type of data, allowing the model to understand complex patterns in both visual and time-based information. Apart from the technical challenges, our goal is to make the predictive model more robust in noisy conditions and perform better than current methods. We also deal with imbalanced datasets and use an uncertainty loss function, yielding improved results while simultaneously providing a principled means of modeling uncertainty. Additionally, we include attention mechanisms to fuse different modalities, allowing the model to focus on what's important for each task. We tested our approach using the comprehensive multimodal MIMIC dataset, combining MIMIC-IV and MIMIC-CXR datasets. Our experiments show that our method is effective in improving multimodal deep learning for clinical applications. The code will be made available online.

Gollam Rabby, Diyana Muhammed, Prasenjit Mitra et al.

Scientific hypothesis generation is a fundamentally challenging task in research, requiring the synthesis of novel and empirically grounded insights. Traditional approaches rely on human intuition and domain expertise, while purely large language model (LLM) based methods often struggle to produce hypotheses that are both innovative and reliable. To address these limitations, we propose the Monte Carlo Nash Equilibrium Self-Refine Tree (MC-NEST), a novel framework that integrates Monte Carlo Tree Search with Nash Equilibrium strategies to iteratively refine and validate hypotheses. MC-NEST dynamically balances exploration and exploitation through adaptive sampling strategies, which prioritize high-potential hypotheses while maintaining diversity in the search space. We demonstrate the effectiveness of MC-NEST through comprehensive experiments across multiple domains, including biomedicine, social science, and computer science. MC-NEST achieves average scores of 2.65, 2.74, and 2.80 (on a 1-3 scale) for novelty, clarity, significance, and verifiability metrics on the social science, computer science, and biomedicine datasets, respectively, outperforming state-of-the-art prompt-based methods, which achieve 2.36, 2.51, and 2.52 on the same datasets. These results underscore MC-NEST's ability to generate high-quality, empirically grounded hypotheses across diverse domains. Furthermore, MC-NEST facilitates structured human-AI collaboration, ensuring that LLMs augment human creativity rather than replace it. By addressing key challenges such as iterative refinement and the exploration-exploitation balance, MC-NEST sets a new benchmark in automated hypothesis generation. Additionally, MC-NEST's ethical design enables responsible AI use, emphasizing transparency and human supervision in hypothesis generation.

Ali Al-Lawati, Jason Lucas, Zhiwei Zhang et al.

In-context learning (ICL) effectively conditions large language models (LLMs) for molecular tasks, such as property prediction and molecule captioning, by embedding carefully selected demonstration examples into the input prompt. This approach avoids the computational overhead of extensive pertaining and fine-tuning. However, current prompt retrieval methods for molecular tasks have relied on molecule feature similarity, such as Morgan fingerprints, which do not adequately capture the global molecular and atom-binding relationships. As a result, these methods fail to represent the full complexity of molecular structures during inference. Moreover, small-to-medium-sized LLMs, which offer simpler deployment requirements in specialized systems, have remained largely unexplored in the molecular ICL literature. To address these gaps, we propose a self-supervised learning technique, GAMIC (Graph-Aligned Molecular In-Context learning, which aligns global molecular structures, represented by graph neural networks (GNNs), with textual captions (descriptions) while leveraging local feature similarity through Morgan fingerprints. In addition, we introduce a Maximum Marginal Relevance (MMR) based diversity heuristic during retrieval to optimize input prompt demonstration samples. Our experimental findings using diverse benchmark datasets show GAMIC outperforms simple Morgan-based ICL retrieval methods across all tasks by up to 45%.

Saptarshi Sengupta, Shreya Ghosh, Prasenjit Mitra et al.

Sentiment Analysis (SA) refers to the task of associating a view polarity (usually, positive, negative, or neutral; or even fine-grained such as slightly angry, sad, etc.) to a given text, essentially breaking it down to a supervised (since we have the view labels apriori) classification task. Although heavily studied in resource-rich languages such as English thus pushing the SOTA by leaps and bounds, owing to the arrival of the Transformer architecture, the same cannot be said for resource-poor languages such as Bengali (BN). For a language spoken by roughly 300 million people, the technology enabling them to run trials on their favored tongue is severely lacking. In this paper, we analyze the SOTA for SA in Bengali, particularly, Transformer-based models. We discuss available datasets, their drawbacks, the nuances associated with Bengali i.e. what makes this a challenging language to apply SA on, and finally provide insights for future direction to mitigate the limitations in the field.

Mahounan Pericles Adjovi, Victor Olufemi, Roald Eiselen et al.

This survey provides a comprehensive catalog of publicly available text and speech resources for two West African languages: Hausa, an Afroasiatic language with approximately 80-100 million speakers, and Fongbe, a Niger-Congo language spoken by approximately 2 million people in Benin. These languages represent contrasting cases on the resource availability spectrum. We address the question: \textit{What is the current state of publicly available NLP resources for Hausa and Fongbe, and what gaps remain?} Through systematic search of academic repositories, data platforms, and web sources, we catalog parallel corpora, monolingual text collections, speech datasets, pre-trained models, and evaluation benchmarks. For each resource, we document size, domain coverage, format, licensing, and accessibility. Our findings reveal that Hausa benefits from broader text resource diversity across news, encyclopedic, and educational domains. Fongbe, while having more limited text resources, has been the focus of recent academic speech data collection initiatives. Both languages are represented in Masakhane benchmarks for NER and POS tagging. We provide task-specific recommendations and identify priority gaps including domain-diverse Fongbe text and dedicated Hausa speech corpora.

Priyanka Kargupta, Nan Zhang, Yunyi Zhang et al. · amazon-science

The rapid evolution of scientific fields introduces challenges in organizing and retrieving scientific literature. While expert-curated taxonomies have traditionally addressed this need, the process is time-consuming and expensive. Furthermore, recent automatic taxonomy construction methods either (1) over-rely on a specific corpus, sacrificing generalizability, or (2) depend heavily on the general knowledge of large language models (LLMs) contained within their pre-training datasets, often overlooking the dynamic nature of evolving scientific domains. Additionally, these approaches fail to account for the multi-faceted nature of scientific literature, where a single research paper may contribute to multiple dimensions (e.g., methodology, new tasks, evaluation metrics, benchmarks). To address these gaps, we propose TaxoAdapt, a framework that dynamically adapts an LLM-generated taxonomy to a given corpus across multiple dimensions. TaxoAdapt performs iterative hierarchical classification, expanding both the taxonomy width and depth based on corpus' topical distribution. We demonstrate its state-of-the-art performance across a diverse set of computer science conferences over the years to showcase its ability to structure and capture the evolution of scientific fields. As a multidimensional method, TaxoAdapt generates taxonomies that are 26.51% more granularity-preserving and 50.41% more coherent than the most competitive baselines judged by LLMs.

Ali Al-Lawati, Elsayed Eshra, Prasenjit Mitra

Trajectory generation is an important concern in pedestrian, vehicle, and wildlife movement studies. Generated trajectories help enrich the training corpus in relation to deep learning applications, and may be used to facilitate simulation tasks. This is especially significant in the wildlife domain, where the cost of obtaining additional real data can be prohibitively expensive, time-consuming, and bear ethical considerations. In this paper, we introduce WildGEN: a conceptual framework that addresses this challenge by employing a Variational Auto-encoders (VAEs) based method for the acquisition of movement characteristics exhibited by wild geese over a long horizon using a sparse set of truth samples. A subsequent post-processing step of the generated trajectories is performed based on smoothing filters to reduce excessive wandering. Our evaluation is conducted through visual inspection and the computation of the Hausdorff distance between the generated and real trajectories. In addition, we utilize the Pearson Correlation Coefficient as a way to measure how realistic the trajectories are based on the similarity of clusters evaluated on the generated and real trajectories.

Chen Wu, Qian Ma, Prasenjit Mitra et al.

Knowledge distillation has become a cornerstone in modern machine learning systems, celebrated for its ability to transfer knowledge from a large, complex teacher model to a more efficient student model. Traditionally, this process is regarded as secure, assuming the teacher model is clean. This belief stems from conventional backdoor attacks relying on poisoned training data with backdoor triggers and attacker-chosen labels, which are not involved in the distillation process. Instead, knowledge distillation uses the outputs of a clean teacher model to guide the student model, inherently preventing recognition or response to backdoor triggers as intended by an attacker. In this paper, we challenge this assumption by introducing a novel attack methodology that strategically poisons the distillation dataset with adversarial examples embedded with backdoor triggers. This technique allows for the stealthy compromise of the student model while maintaining the integrity of the teacher model. Our innovative approach represents the first successful exploitation of vulnerabilities within the knowledge distillation process using clean teacher models. Through extensive experiments conducted across various datasets and attack settings, we demonstrate the robustness, stealthiness, and effectiveness of our method. Our findings reveal previously unrecognized vulnerabilities and pave the way for future research aimed at securing knowledge distillation processes against backdoor attacks.

Farhana Keya, Gollam Rabby, Prasenjit Mitra et al.

Every scientific discovery starts with an idea inspired by prior work, interdisciplinary concepts, and emerging challenges. Recent advancements in large language models (LLMs) trained on scientific corpora have driven interest in AI-supported idea generation. However, generating context-aware, high-quality, and innovative ideas remains challenging. We introduce SCI-IDEA, a framework that uses LLM prompting strategies and Aha Moment detection for iterative idea refinement. SCI-IDEA extracts essential facets from research publications, assessing generated ideas on novelty, excitement, feasibility, and effectiveness. Comprehensive experiments validate SCI-IDEA's effectiveness, achieving average scores of 6.84, 6.86, 6.89, and 6.84 (on a 1-10 scale) across novelty, excitement, feasibility, and effectiveness, respectively. Evaluations employed GPT-4o, GPT-4.5, DeepSeek-32B (each under 2-shot prompting), and DeepSeek-70B (3-shot prompting), with token-level embeddings used for Aha Moment detection. Similarly, it achieves scores of 6.87, 6.86, 6.83, and 6.87 using GPT-4o under 5-shot prompting, GPT-4.5 under 3-shot prompting, DeepSeek-32B under zero-shot chain-of-thought prompting, and DeepSeek-70B under 5-shot prompting with sentence-level embeddings. We also address ethical considerations such as intellectual credit, potential misuse, and balancing human creativity with AI-driven ideation. Our results highlight SCI-IDEA's potential to facilitate the structured and flexible exploration of context-aware scientific ideas, supporting innovation while maintaining ethical standards.

Nan Zhang, Eugene Kwek, Yusen Zhang et al.

Compression methods, including quantization, distillation, and pruning, improve the computational efficiency of large reasoning models (LRMs). However, existing studies either fail to sufficiently compare all three compression methods on LRMs or lack in-depth interpretation analysis. In this paper, we investigate how the reasoning capabilities of LRMs are compromised during compression, through performance benchmarking and mechanistic interpretation. To uncover the effects of compression on reasoning performance, we benchmark quantized, distilled, and pruned DeepSeek-R1 models on four reasoning datasets (AIME 2024, FOLIO, Temporal Sequences, and MuSiQue). To precisely locate compression effects on model weights, we adapt difference of means and attribution patching techniques, focusing on the activation of every linear component in compressed LRMs, to interpret fine-grained causal relationships between weights and various reasoning capabilities. This fine-grained interpretation addresses a fundamental question of compression: which weights are the most important for reasoning? Overall, we find dynamically quantized 2.51-bit R1 reaches close-to-R1 performance. With empirical verification, we present three main findings that generalize across both Llama and Qwen: (1) Weight count has a greater impact on LRMs' knowledge memorization than reasoning, highlighting the risks of pruning and distillation; (2) The MLP up projection in the final layer of distilled LRMs is one of the most important components, offering a new perspective on locating critical weights - a fundamental problem in model compression; and (3) Current quantization methods overly compress the final-layer modules and MLP gate projections, so protecting just 2% of all weights that are excessively compressed can raise average accuracy by 6.57%, greatly surpassing the state-of-the-art.

Denis Musinguzi, Caren Han, Prasenjit Mitra

Unlike conventional single-image models, differential medical VQA frameworks process multiple images to identify differences, mirroring the comparative diagnostic workflow of radiologists. However, standard vision encoders trained on contrastive or classification objectives often fail to capture the subtle visual variations necessary for distinguishing disease progression from acquisition differences. To address this limitation, we introduce a pretraining framework that incorporates location-aware tasks, including automatic referring expressions (AREF), grounded captioning (GCAP), and conditional automatic referring expressions (CAREF). These specific tasks enable the vision encoder to learn fine-grained, spatially grounded visual representations that are often overlooked by traditional pre-training methods. We subsequently integrate this enhanced vision encoder with a language model to perform medical difference VQA. Experimental results demonstrate that our approach achieves state-of-the-art performance in detecting and reasoning about clinically relevant changes in chest X-ray images.

Tianrong Zhang, Zhaohan Xi, Ting Wang et al.

Pre-trained language models (PLMs) have attracted enormous attention over the past few years with their unparalleled performances. Meanwhile, the soaring cost to train PLMs as well as their amazing generalizability have jointly contributed to few-shot fine-tuning and prompting as the most popular training paradigms for natural language processing (NLP) models. Nevertheless, existing studies have shown that these NLP models can be backdoored such that model behavior is manipulated when trigger tokens are presented. In this paper, we propose PromptFix, a novel backdoor mitigation strategy for NLP models via adversarial prompt-tuning in few-shot settings. Unlike existing NLP backdoor removal methods, which rely on accurate trigger inversion and subsequent model fine-tuning, PromptFix keeps the model parameters intact and only utilizes two extra sets of soft tokens which approximate the trigger and counteract it respectively. The use of soft tokens and adversarial optimization eliminates the need to enumerate possible backdoor configurations and enables an adaptive balance between trigger finding and preservation of performance. Experiments with various backdoor attacks validate the effectiveness of the proposed method and the performances when domain shift is present further shows PromptFix's applicability to models pretrained on unknown data source which is the common case in prompt tuning scenarios.

Scott Pezanowski, Prasenjit Mitra

Analyzing the geographic movement of humans, animals, and other phenomena is a growing field of research. This research has benefited urban planning, logistics, animal migration understanding, and much more. Typically, the movement is captured as precise geographic coordinates and time stamps with Global Positioning Systems (GPS). Although some research uses computational techniques to take advantage of implicit movement in descriptions of route directions, hiking paths, and historical exploration routes, innovation would accelerate with a large and diverse corpus. We created a corpus of sentences labeled as describing geographic movement or not and including the type of entity moving. Creating this corpus proved difficult without any comparable corpora to start with, high human labeling costs, and since movement can at times be interpreted differently. To overcome these challenges, we developed an iterative process employing hand labeling, crowd voting for confirmation, and machine learning to predict more labels. By merging advances in word embeddings with traditional machine learning models and model ensembling, prediction accuracy is at an acceptable level to produce a large silver-standard corpus despite the small gold-standard corpus training set. Our corpus will likely benefit computational processing of geography in text and spatial cognition, in addition to detection of movement.

Chen Wu, Sencun Zhu, Prasenjit Mitra

Federated Learning (FL) is designed to protect the data privacy of each client during the training process by transmitting only models instead of the original data. However, the trained model may memorize certain information about the training data. With the recent legislation on right to be forgotten, it is crucially essential for the FL model to possess the ability to forget what it has learned from each client. We propose a novel federated unlearning method to eliminate a client's contribution by subtracting the accumulated historical updates from the model and leveraging the knowledge distillation method to restore the model's performance without using any data from the clients. This method does not have any restrictions on the type of neural networks and does not rely on clients' participation, so it is practical and efficient in the FL system. We further introduce backdoor attacks in the training process to help evaluate the unlearning effect. Experiments on three canonical datasets demonstrate the effectiveness and efficiency of our method.

Scott Pezanowski, Alan M. MacEachren, Prasenjit Mitra

Understanding movement described in text documents is important since text descriptions of movement contain a wealth of geographic and contextual information about the movement of people, wildlife, goods, and much more. Our research makes several contributions to improve our understanding of movement descriptions in text. First, we show how interpreting geographic movement described in text is challenging because of general spatial terms, linguistic constructions that make the thing(s) moving unclear, and many types of temporal references and groupings, among others. Next, as a step to overcome these challenges, we report on an experiment with human subjects through which we identify multiple important characteristics of movement descriptions (found in text) that humans use to differentiate one movement description from another. Based on our empirical results, we provide recommendations for computational analysis using movement described in text documents. Our findings contribute towards an improved understanding of the important characteristics of the underused information about geographic movement that is in the form of text descriptions.

Gregory Glatzer, Prasenjit Mitra, Johnson Kinyua

Understanding the movement of animals is crucial to conservation efforts. Past research often focuses on factors affecting movement, rather than locations of interest that animals return to or habitat. We explore the use of clustering to identify locations of interest to African Elephants in regions of Sub-Saharan Africa. Our analysis was performed using publicly available datasets for tracking African elephants at Kruger National Park (KNP), South Africa; Etosha National Park, Namibia; as well as areas in Burkina Faso and the Congo. Using the DBSCAN and KMeans clustering algorithms, we calculate clusters and centroids to simplify elephant movement data and highlight important locations of interest. Through a comparison of feature spaces with and without temperature, we show that temperature is an important feature to explain movement clustering. Recognizing the importance of temperature, we develop a technique to add external temperature data from an API to other geospatial datasets that would otherwise not have temperature data. After addressing the hurdles of using external data with marginally different timestamps, we consider the quality of this data, and the quality of the centroids of the clusters calculated based on this external temperature data. Finally, we overlay these centroids onto satellite imagery and locations of human settlements to validate the real-life applications of the calculated centroids to identify locations of interest for elephants. As expected, we confirmed that elephants tend to cluster their movement around sources of water as well as some human settlements, especially those with water holes. Identifying key locations of interest for elephants is beneficial in predicting the movement of elephants and preventing poaching. These methods may in the future be applied to other animals beyond elephants to identify locations of interests for them.

Connor Heaton, Prasenjit Mitra

Major League Baseball (MLB) has a storied history of using statistics to better understand and discuss the game of baseball, with an entire discipline of statistics dedicated to the craft, known as sabermetrics. At their core, all sabermetrics seek to quantify some aspect of the game, often a specific aspect of a player's skill set - such as a batter's ability to drive in runs (RBI) or a pitcher's ability to keep batters from reaching base (WHIP). While useful, such statistics are fundamentally limited by the fact that they are derived from an account of what happened on the field, not how it happened. As a first step towards alleviating this shortcoming, we present a novel, contrastive learning-based framework for describing player form in the MLB. We use form to refer to the way in which a player has impacted the course of play in their recent appearances. Concretely, a player's form is described by a 72-dimensional vector. By comparing clusters of players resulting from our form representations and those resulting from traditional abermetrics, we demonstrate that our form representations contain information about how players impact the course of play, not present in traditional, publicly available statistics. We believe these embeddings could be utilized to predict both in-game and game-level events, such as the result of an at-bat or the winner of a game.

Athar Sefid, Jian Wu, Prasenjit Mitra et al.

Presentation slides describing the content of scientific and technical papers are an efficient and effective way to present that work. However, manually generating presentation slides is labor intensive. We propose a method to automatically generate slides for scientific papers based on a corpus of 5000 paper-slide pairs compiled from conference proceedings websites. The sentence labeling module of our method is based on SummaRuNNer, a neural sequence model for extractive summarization. Instead of ranking sentences based on semantic similarities in the whole document, our algorithm measures importance and novelty of sentences by combining semantic and lexical features within a sentence window. Our method outperforms several baseline methods including SummaRuNNer by a significant margin in terms of ROUGE score.

Chen Wu, Xian Yang, Sencun Zhu et al.

Malicious clients can attack federated learning systems using malicious data, including backdoor samples, during the training phase. The compromised global model will perform well on the validation dataset designed for the task, but a small subset of data with backdoor patterns may trigger the model to make a wrong prediction. There has been an arms race between attackers who tried to conceal attacks and defenders who tried to detect attacks during the aggregation stage of training on the server-side. In this work, we propose a new and effective method to mitigate backdoor attacks after the training phase. Specifically, we design a federated pruning method to remove redundant neurons in the network and then adjust the model's extreme weight values. Our experiments conducted on distributed Fashion-MNIST show that our method can reduce the average attack success rate from 99.7% to 1.9% with a 5.5% loss of test accuracy on the validation dataset. To minimize the pruning influence on test accuracy, we can fine-tune after pruning, and the attack success rate drops to 6.4%, with only a 1.7% loss of test accuracy. Further experiments under Distributed Backdoor Attacks on CIFAR-10 also show promising results that the average attack success rate drops more than 70% with less than 2% loss of test accuracy on the validation dataset.

Connor T. Heaton, Prasenjit Mitra

The novel coronavirus disease 2019 (COVID-19) began in Wuhan, China in late 2019 and to date has infected over 14M people worldwide, resulting in over 750,000 deaths. On March 10, 2020 the World Health Organization (WHO) declared the outbreak a global pandemic. Many academics and researchers, not restricted to the medical domain, began publishing papers describing new discoveries. However, with the large influx of publications, it was hard for these individuals to sift through the large amount of data and make sense of the findings. The White House and a group of industry research labs, lead by the Allen Institute for AI, aggregated over 200,000 journal articles related to a variety of coronaviruses and tasked the community with answering key questions related to the corpus, releasing the dataset as CORD-19. The information retrieval (IR) community repurposed the journal articles within CORD-19 to more closely resemble a classic TREC-style competition, dubbed TREC-COVID, with human annotators providing relevancy judgements at the end of each round of competition. Seeing the related endeavors, we set out to repurpose the relevancy annotations for TREC-COVID tasks to identify journal articles in CORD-19 which are relevant to the key questions posed by CORD-19. A BioBERT model trained on this repurposed dataset prescribes relevancy annotations for CORD-19 tasks that have an overall agreement of 0.4430 with majority human annotations in terms of Cohen's kappa. We present the methodology used to construct the new dataset and describe the decision process used throughout.

Athar Sefid, Clyde Lee Giles, Prasenjit Mitra

We introduce an extractive method that will summarize long scientific papers. Our model uses presentation slides provided by the authors of the papers as the gold summary standard to label the sentences. The sentences are ranked based on their novelty and their importance as estimated by deep neural networks. Our window-based extractive labeling of sentences results in the improvement of at least 4 ROUGE1-Recall points.

Xianfeng Tang, Huaxiu Yao, Yiwei Sun et al.

Graph Convolutional Networks (GCNs) show promising results for semi-supervised learning tasks on graphs, thus become favorable comparing with other approaches. Despite the remarkable success of GCNs, it is difficult to train GCNs with insufficient supervision. When labeled data are limited, the performance of GCNs becomes unsatisfying for low-degree nodes. While some prior work analyze successes and failures of GCNs on the entire model level, profiling GCNs on individual node level is still underexplored. In this paper, we analyze GCNs in regard to the node degree distribution. From empirical observation to theoretical proof, we confirm that GCNs are biased towards nodes with larger degrees with higher accuracy on them, even if high-degree nodes are underrepresented in most graphs. We further develop a novel Self-Supervised-Learning Degree-Specific GCN (SL-DSGC) that mitigate the degree-related biases of GCNs from model and data aspects. Firstly, we propose a degree-specific GCN layer that captures both discrepancies and similarities of nodes with different degrees, which reduces the inner model-aspect biases of GCNs caused by sharing the same parameters with all nodes. Secondly, we design a self-supervised-learning algorithm that creates pseudo labels with uncertainty scores on unlabeled nodes with a Bayesian neural network. Pseudo labels increase the chance of connecting to labeled neighbors for low-degree nodes, thus reducing the biases of GCNs from the data perspective. Uncertainty scores are further exploited to weight pseudo labels dynamically in the stochastic gradient descent for SL-DSGC. Experiments on three benchmark datasets show SL-DSGC not only outperforms state-of-the-art self-training/self-supervised-learning GCN methods, but also improves GCN accuracy dramatically for low-degree nodes.

Xianfeng Tang, Yozen Liu, Neil Shah et al.

With the rapid growth and prevalence of social network applications (Apps) in recent years, understanding user engagement has become increasingly important, to provide useful insights for future App design and development. While several promising neural modeling approaches were recently pioneered for accurate user engagement prediction, their black-box designs are unfortunately limited in model explainability. In this paper, we study a novel problem of explainable user engagement prediction for social network Apps. First, we propose a flexible definition of user engagement for various business scenarios, based on future metric expectations. Next, we design an end-to-end neural framework, FATE, which incorporates three key factors that we identify to influence user engagement, namely friendships, user actions, and temporal dynamics to achieve explainable engagement predictions. FATE is based on a tensor-based graph neural network (GNN), LSTM and a mixture attention mechanism, which allows for (a) predictive explanations based on learned weights across different feature categories, (b) reduced network complexity, and (c) improved performance in both prediction accuracy and training/inference time. We conduct extensive experiments on two large-scale datasets from Snapchat, where FATE outperforms state-of-the-art approaches by ${\approx}10\%$ error and ${\approx}20\%$ runtime reduction. We also evaluate explanations from FATE, showing strong quantitative and qualitative performance.

Xianfeng Tang, Huaxiu Yao, Yiwei Sun et al.

Multivariate time series (MTS) forecasting is widely used in various domains, such as meteorology and traffic. Due to limitations on data collection, transmission, and storage, real-world MTS data usually contains missing values, making it infeasible to apply existing MTS forecasting models such as linear regression and recurrent neural networks. Though many efforts have been devoted to this problem, most of them solely rely on local dependencies for imputing missing values, which ignores global temporal dynamics. Local dependencies/patterns would become less useful when the missing ratio is high, or the data have consecutive missing values; while exploring global patterns can alleviate such problems. Thus, jointly modeling local and global temporal dynamics is very promising for MTS forecasting with missing values. However, work in this direction is rather limited. Therefore, we study a novel problem of MTS forecasting with missing values by jointly exploring local and global temporal dynamics. We propose a new framework LGnet, which leverages memory network to explore global patterns given estimations from local perspectives. We further introduce adversarial training to enhance the modeling of global temporal distribution. Experimental results on real-world datasets show the effectiveness of LGnet for MTS forecasting with missing values and its robustness under various missing ratios.

Rajeev Bhatt Ambati, Saptarashmi Bandyopadhyay, Prasenjit Mitra

Traditional sequence-to-sequence (seq2seq) models and other variations of the attention-mechanism such as hierarchical attention have been applied to the text summarization problem. Though there is a hierarchy in the way humans use language by forming paragraphs from sentences and sentences from words, hierarchical models have usually not worked that much better than their traditional seq2seq counterparts. This effect is mainly because either the hierarchical attention mechanisms are too sparse using hard attention or noisy using soft attention. In this paper, we propose a method based on extracting the highlights of a document; a key concept that is conveyed in a few sentences. In a typical text summarization dataset consisting of documents that are 800 tokens in length (average), capturing long-term dependencies is very important, e.g., the last sentence can be grouped with the first sentence of a document to form a summary. LSTMs (Long Short-Term Memory) proved useful for machine translation. However, they often fail to capture long-term dependencies while modeling long sequences. To address these issues, we have adapted Neural Semantic Encoders (NSE) to text summarization, a class of memory-augmented neural networks by improving its functionalities and proposed a novel hierarchical NSE that outperforms similar previous models significantly. The quality of summarization was improved by augmenting linguistic factors, namely lemma, and Part-of-Speech (PoS) tags, to each word in the dataset for improved vocabulary coverage and generalization. The hierarchical NSE model on factored dataset outperformed the state-of-the-art by nearly 4 ROUGE points. We further designed and used the first GPU-based self-critical Reinforcement Learning model.

Athar Sefid, Jian Wu, Allen C. Ge et al.

Automatically extracted metadata from scholarly documents in PDF formats is usually noisy and heterogeneous, often containing incomplete fields and erroneous values. One common way of cleaning metadata is to use a bibliographic reference dataset. The challenge is to match records between corpora with high precision. The existing solution which is based on information retrieval and string similarity on titles works well only if the titles are cleaned. We introduce a system designed to match scholarly document entities with noisy metadata against a reference dataset. The blocking function uses the classic BM25 algorithm to find the matching candidates from the reference data that has been indexed by ElasticSearch. The core components use supervised methods which combine features extracted from all available metadata fields. The system also leverages available citation information to match entities. The combination of metadata and citation achieves high accuracy that significantly outperforms the baseline method on the same test dataset. We apply this system to match the database of CiteSeerX against Web of Science, PubMed, and DBLP. This method will be deployed in the CiteSeerX system to clean metadata and link records to other scholarly big datasets.

Agnese Chiatti, Dolzodmaa Davaasuren, Nilam Ram et al.

A significant proportion of individuals' daily activities is experienced through digital devices. Smartphones in particular have become one of the preferred interfaces for content consumption and social interaction. Identifying the content embedded in frequently-captured smartphone screenshots is thus a crucial prerequisite to studies of media behavior and health intervention planning that analyze activity interplay and content switching over time. Screenshot images can depict heterogeneous contents and applications, making the a priori definition of adequate taxonomies a cumbersome task, even for humans. Privacy protection of the sensitive data captured on screens means the costs associated with manual annotation are large, as the effort cannot be crowd-sourced. Thus, there is need to examine utility of unsupervised and semi-supervised methods for digital screenshot classification. This work introduces the implications of applying clustering on large screenshot sets when only a limited amount of labels is available. In this paper we develop a framework for combining K-Means clustering with Active Learning for efficient leveraging of labeled and unlabeled samples, with the goal of discovering latent classes and describing a large collection of screenshot data. We tested whether SVM-embedded or XGBoost-embedded solutions for class probability propagation provide for more well-formed cluster configurations. Visual and textual vector representations of the screenshot images are derived and combined to assess the relative contribution of multi-modal features to the overall performance.