Tanwi Mallick

Weiheng Zhong, Tanwi Mallick, Hadi Meidani et al.

Accurate traffic forecasting is vital to an intelligent transportation system. Although many deep learning models have achieved state-of-art performance for short-term traffic forecasting of up to 1 hour, long-term traffic forecasting that spans multiple hours remains a major challenge. Moreover, most of the existing deep learning traffic forecasting models are black box, presenting additional challenges related to explainability and interpretability. We develop Graph Pyramid Autoformer (X-GPA), an explainable attention-based spatial-temporal graph neural network that uses a novel pyramid autocorrelation attention mechanism. It enables learning from long temporal sequences on graphs and improves long-term traffic forecasting accuracy. Our model can achieve up to 35 % better long-term traffic forecast accuracy than that of several state-of-the-art methods. The attention-based scores from the X-GPA model provide spatial and temporal explanations based on the traffic dynamics, which change for normal vs. peak-hour traffic and weekday vs. weekend traffic.

Tanwi Mallick, Prasanna Balaprakash, Jane Macfarlane

Deep-learning-based data-driven forecasting methods have produced impressive results for traffic forecasting. A major limitation of these methods, however, is that they provide forecasts without estimates of uncertainty, which are critical for real-time deployments. We focus on a diffusion convolutional recurrent neural network (DCRNN), a state-of-the-art method for short-term traffic forecasting. We develop a scalable deep ensemble approach to quantify uncertainties for DCRNN. Our approach uses a scalable Bayesian optimization method to perform hyperparameter optimization, selects a set of high-performing configurations, fits a generative model to capture the joint distributions of the hyperparameter configurations, and trains an ensemble of models by sampling a new set of hyperparameter configurations from the generative model. We demonstrate the efficacy of the proposed methods by comparing them with other uncertainty estimation techniques. We show that our generic and scalable approach outperforms the current state-of-the-art Bayesian and a number of other commonly used frequentist techniques.

Tanwi Mallick, Joshua David Bergerson, Duane R. Verner et al.

Natural language processing (NLP) is a promising approach for analyzing large volumes of climate-change and infrastructure-related scientific literature. However, best-in-practice NLP techniques require large collections of relevant documents (corpus). Furthermore, NLP techniques using machine learning and deep learning techniques require labels grouping the articles based on user-defined criteria for a significant subset of a corpus in order to train the supervised model. Even labeling a few hundred documents with human subject-matter experts is a time-consuming process. To expedite this process, we developed a weak supervision-based NLP approach that leverages semantic similarity between categories and documents to (i) establish a topic-specific corpus by subsetting a large-scale open-access corpus and (ii) generate category labels for the topic-specific corpus. In comparison with a months-long process of subject-matter expert labeling, we assign category labels to the whole corpus using weak supervision and supervised learning in about 13 hours. The labeled climate and NCF corpus enable targeted, efficient identification of documents discussing a topic (or combination of topics) of interest and identification of various effects of climate change on critical infrastructure, improving the usability of scientific literature and ultimately supporting enhanced policy and decision making. To demonstrate this capability, we conduct topic modeling on pairs of climate hazards and NCFs to discover trending topics at the intersection of these categories. This method is useful for analysts and decision-makers to quickly grasp the relevant topics and most important documents linked to the topic.

Yangxinyu Xie, Tanwi Mallick

Spatiotemporal graph neural networks have achieved state-of-the-art performance in traffic forecasting. However, they often struggle to forecast congestion accurately due to the limitations of traditional loss functions. While accurate forecasting of regular traffic conditions is crucial, a reliable AI system must also accurately forecast congestion scenarios to maintain safe and efficient transportation. In this paper, we explore various loss functions inspired by heavy tail analysis and imbalanced classification problems to address this issue. We evaluate the efficacy of these loss functions in forecasting traffic speed, with an emphasis on congestion scenarios. Through extensive experiments on real-world traffic datasets, we discovered that when optimizing for Mean Absolute Error (MAE), the MAE-Focal Loss function stands out as the most effective. When optimizing Mean Squared Error (MSE), Gumbel Loss proves to be the superior choice. These choices effectively forecast traffic congestion events without compromising the accuracy of regular traffic speed forecasts. This research enhances deep learning models' capabilities in forecasting sudden speed changes due to congestion and underscores the need for more research in this direction. By elevating the accuracy of congestion forecasting, we advocate for AI systems that are reliable, secure, and resilient in practical traffic management scenarios.

Chih-Hsuan Yang, Tanwi Mallick, Le Chen et al.

Large Language Models (LLMs) in multi-agent systems (MAS) have shown promise for complex tasks, yet current training methods lack principled ways to connect system-level evaluation with agent-level and message-level learning. We propose a theoretical framework that unifies cooperative game-theoretic attribution with process reward modeling to transform system evaluation into agent credit and then into response-level signals. Unlike prior approaches that rely only on attribution (e.g., Shapley) or step-level labels (e.g., PRM), our method produces local, signed, and credit-conserving signals. In success cases, Shapley-based credit assignment fairly allocates outcomes across agents and is refined into per-message rewards that promote cooperation while discouraging redundancy or sabotage. In failure cases, first-error localization yields repair-aware preferences that penalize harmful steps while rewarding corrective attempts. The resulting signals are bounded, cooperative, and directly compatible with reinforcement-based or preference-based post-training, providing a unified and auditable pathway from global evaluation to local supervision in LLM multi-agent training. Our contribution is conceptual: we present a theoretical foundation and training signals, leaving empirical validation for future work.

Bowen Jiang, Yangxinyu Xie, Zhuoqun Hao et al.

This study introduces a hypothesis-testing framework to assess whether large language models (LLMs) possess genuine reasoning abilities or primarily depend on token bias. We go beyond evaluating LLMs on accuracy; rather, we aim to investigate their token bias in solving logical reasoning tasks. Specifically, we develop carefully controlled synthetic datasets, featuring conjunction fallacy and syllogistic problems. Our framework outlines a list of hypotheses where token biases are readily identifiable, with all null hypotheses assuming genuine reasoning capabilities of LLMs. The findings in this study suggest, with statistical guarantee, that most LLMs still struggle with logical reasoning. While they may perform well on classic problems, their success largely depends on recognizing superficial patterns with strong token bias, thereby raising concerns about their actual reasoning and generalization abilities. Codes and data are open-sourced at https://github.com/bowen-upenn/llm_token_bias.

Yangxinyu Xie, Bowen Jiang, Tanwi Mallick et al.

Recent advancement of large language models (LLMs) represents a transformational capability at the frontier of artificial intelligence. However, LLMs are generalized models, trained on extensive text corpus, and often struggle to provide context-specific information, particularly in areas requiring specialized knowledge, such as wildfire details within the broader context of climate change. For decision-makers focused on wildfire resilience and adaptation, it is crucial to obtain responses that are not only precise but also domain-specific. To that end, we developed WildfireGPT, a prototype LLM agent designed to transform user queries into actionable insights on wildfire risks. We enrich WildfireGPT by providing additional context, such as climate projections and scientific literature, to ensure its information is current, relevant, and scientifically accurate. This enables WildfireGPT to be an effective tool for delivering detailed, user-specific insights on wildfire risks to support a diverse set of end users, including but not limited to researchers and engineers, for making positive impact and decision making.

Tanwi Mallick, John Murphy, Joshua David Bergerson et al.

Understanding the multifaceted effects of climate change across diverse geographic locations is crucial for timely adaptation and the development of effective mitigation strategies. As the volume of scientific literature on this topic continues to grow exponentially, manually reviewing these documents has become an immensely challenging task. Utilizing Natural Language Processing (NLP) techniques to analyze this wealth of information presents an efficient and scalable solution. By gathering extensive amounts of peer-reviewed articles and studies, we can extract and process critical information about the effects of climate change in specific regions. We employ BERT (Bidirectional Encoder Representations from Transformers) for Named Entity Recognition (NER), which enables us to efficiently identify specific geographies within the climate literature. This, in turn, facilitates location-specific analyses. We conduct region-specific climate trend analyses to pinpoint the predominant themes or concerns related to climate change within a particular area, trace the temporal progression of these identified issues, and evaluate their frequency, severity, and potential development over time. These in-depth examinations of location-specific climate data enable the creation of more customized policy-making, adaptation, and mitigation strategies, addressing each region's unique challenges and providing more effective solutions rooted in data-driven insights. This approach, founded on a thorough exploration of scientific texts, offers actionable insights to a wide range of stakeholders, from policymakers to engineers to environmentalists. By proactively understanding these impacts, societies are better positioned to prepare, allocate resources wisely, and design tailored strategies to cope with future climate conditions, ensuring a more resilient future for all.

Yangxinyu Xie, Bowen Jiang, Tanwi Mallick et al.

Large language models (LLMs) are a transformational capability at the frontier of artificial intelligence and machine learning that can support decision-makers in addressing pressing societal challenges such as extreme natural hazard events. As generalized models, LLMs often struggle to provide context-specific information, particularly in areas requiring specialized knowledge. In this work we propose a retrieval-augmented generation (RAG)-based multi-agent LLM system to support analysis and decision-making in the context of natural hazards and extreme weather events. As a proof of concept, we present WildfireGPT, a specialized system focused on wildfire hazards. The architecture employs a user-centered, multi-agent design to deliver tailored risk insights across diverse stakeholder groups. By integrating natural hazard and extreme weather projection data, observational datasets, and scientific literature through an RAG framework, the system ensures both the accuracy and contextual relevance of the information it provides. Evaluation across ten expert-led case studies demonstrates that WildfireGPT significantly outperforms existing LLM-based solutions for decision support.

Qipeng Qian, Tanwi Mallick

Traffic forecasting is the foundation for intelligent transportation systems. Spatiotemporal graph neural networks have demonstrated state-of-the-art performance in traffic forecasting. However, these methods do not explicitly model some of the natural characteristics in traffic data, such as the multiscale structure that encompasses spatial and temporal variations at different levels of granularity or scale. To that end, we propose a Wavelet-Inspired Graph Convolutional Recurrent Network (WavGCRN) which combines multiscale analysis (MSA)-based method with Deep Learning (DL)-based method. In WavGCRN, the traffic data is decomposed into time-frequency components with Discrete Wavelet Transformation (DWT), constructing a multi-stream input structure; then Graph Convolutional Recurrent networks (GCRNs) are employed as encoders for each stream, extracting spatiotemporal features in different scales; and finally the learnable Inversed DWT and GCRN are combined as the decoder, fusing the information from all streams for traffic metrics reconstruction and prediction. Furthermore, road-network-informed graphs and data-driven graph learning are combined to accurately capture spatial correlation. The proposed method can offer well-defined interpretability, powerful learning capability, and competitive forecasting performance on real-world traffic data sets.

Franck Cappello, Sandeep Madireddy, Robert Underwood et al.

Recent advancements have positioned AI, and particularly Large Language Models (LLMs), as transformative tools for scientific research, capable of addressing complex tasks that require reasoning, problem-solving, and decision-making. Their exceptional capabilities suggest their potential as scientific research assistants but also highlight the need for holistic, rigorous, and domain-specific evaluation to assess effectiveness in real-world scientific applications. This paper describes a multifaceted methodology for Evaluating AI models as scientific Research Assistants (EAIRA) developed at Argonne National Laboratory. This methodology incorporates four primary classes of evaluations. 1) Multiple Choice Questions to assess factual recall; 2) Open Response to evaluate advanced reasoning and problem-solving skills; 3) Lab-Style Experiments involving detailed analysis of capabilities as research assistants in controlled environments; and 4) Field-Style Experiments to capture researcher-LLM interactions at scale in a wide range of scientific domains and applications. These complementary methods enable a comprehensive analysis of LLM strengths and weaknesses with respect to their scientific knowledge, reasoning abilities, and adaptability. Recognizing the rapid pace of LLM advancements, we designed the methodology to evolve and adapt so as to ensure its continued relevance and applicability. This paper describes the methodology state at the end of February 2025. Although developed within a subset of scientific domains, the methodology is designed to be generalizable to a wide range of scientific domains.

Yangxinyu Xie, Tao Wang, Soham Mallick et al.

While LLMs have seen substantial improvement in reasoning capabilities, they also sometimes overthink, generating unnecessary reasoning steps, particularly under uncertainty, given ill-posed or ambiguous queries. We introduce statistically principled early stopping methods that monitor uncertainty signals during generation to mitigate this issue. Our first approach is parametric: it models inter-arrival times of uncertainty keywords as a renewal process and applies sequential testing for stopping. Our second approach is nonparametric and provides finite-sample guarantees on the probability of halting too early on well-posed queries. We conduct empirical evaluations on reasoning tasks across several domains and models. Our results indicate that uncertainty-aware early stopping can improve both efficiency and reliability in LLM reasoning, and we observe especially significant gains for math reasoning.

Bowen Jiang, Yangxinyu Xie, Xiaomeng Wang et al.

We present GeoGrid-Bench, a benchmark designed to evaluate the ability of foundation models to understand geo-spatial data in the grid structure. Geo-spatial datasets pose distinct challenges due to their dense numerical values, strong spatial and temporal dependencies, and unique multimodal representations including tabular data, heatmaps, and geographic visualizations. To assess how foundation models can support scientific research in this domain, GeoGrid-Bench features large-scale, real-world data covering 16 climate variables across 150 locations and extended time frames. The benchmark includes approximately 3,200 question-answer pairs, systematically generated from 8 domain expert-curated templates to reflect practical tasks encountered by human scientists. These range from basic queries at a single location and time to complex spatiotemporal comparisons across regions and periods. Our evaluation reveals that vision-language models perform best overall, and we provide a fine-grained analysis of the strengths and limitations of different foundation models in different geo-spatial tasks. This benchmark offers clearer insights into how foundation models can be effectively applied to geo-spatial data analysis and used to support scientific research.

Siddeshwar Raghavan, Tanwi Mallick

We present MOSAIC, a multi-agent Large Language Model (LLM) framework for solving challenging scientific coding tasks. Unlike general-purpose coding, scientific workflows require algorithms that are rigorous, interconnected with deep domain knowledge, and incorporate domain-specific reasoning, as well as algorithm iteration without requiring I/O test cases. Many scientific problems also require a sequence of subproblems to be solved, leading to the final desired result. MOSAIC is designed as a training-free framework with specially designed agents to self-reflect, create the rationale, code, and debug within a student-teacher paradigm to address the challenges of scientific code generation. This design facilitates stepwise problem decomposition, targeted error correction, and, when combined with our Consolidated Context Window (CCW), mitigates LLM hallucinations when solving complex scientific tasks involving chained subproblems. We evaluate MOSAIC on scientific coding benchmarks and demonstrate that our specialized agentic framework outperforms existing approaches in terms of accuracy, robustness, and interpretability.

Yixuan Sun, Tanwi Mallick, Prasanna Balaprakash et al.

Using the data from loop detector sensors for near-real-time detection of traffic incidents in highways is crucial to averting major traffic congestion. While recent supervised machine learning methods offer solutions to incident detection by leveraging human-labeled incident data, the false alarm rate is often too high to be used in practice. Specifically, the inconsistency in the human labeling of the incidents significantly affects the performance of supervised learning models. To that end, we focus on a data-centric approach to improve the accuracy and reduce the false alarm rate of traffic incident detection on highways. We develop a weak supervised learning workflow to generate high-quality training labels for the incident data without the ground truth labels, and we use those generated labels in the supervised learning setup for final detection. This approach comprises three stages. First, we introduce a data preprocessing and curation pipeline that processes traffic sensor data to generate high-quality training data through leveraging labeling functions, which can be domain knowledge-related or simple heuristic rules. Second, we evaluate the training data generated by weak supervision using three supervised learning models -- random forest, k-nearest neighbors, and a support vector machine ensemble -- and long short-term memory classifiers. The results show that the accuracy of all of the models improves significantly after using the training data generated by weak supervision. Third, we develop an online real-time incident detection approach that leverages the model ensemble and the uncertainty quantification while detecting incidents. Overall, we show that our proposed weak supervised learning workflow achieves a high incident detection rate (0.90) and low false alarm rate (0.08).

Tanwi Mallick, Mariam Kiran, Bashir Mohammed et al.

Wide area networking infrastructures (WANs), particularly science and research WANs, are the backbone for moving large volumes of scientific data between experimental facilities and data centers. With demands growing at exponential rates, these networks are struggling to cope with large data volumes, real-time responses, and overall network performance. Network operators are increasingly looking for innovative ways to manage the limited underlying network resources. Forecasting network traffic is a critical capability for proactive resource management, congestion mitigation, and dedicated transfer provisioning. To this end, we propose a nonautoregressive graph-based neural network for multistep network traffic forecasting. Specifically, we develop a dynamic variant of diffusion convolutional recurrent neural networks to forecast traffic in research WANs. We evaluate the efficacy of our approach on real traffic from ESnet, the U.S. Department of Energy's dedicated science network. Our results show that compared to classical forecasting methods, our approach explicitly learns the dynamic nature of spatiotemporal traffic patterns, showing significant improvements in forecasting accuracy. Our technique can surpass existing statistical and deep learning approaches by achieving approximately 20% mean absolute percentage error for multiple hours of forecasts despite dynamic network traffic settings.

Tanwi Mallick, Patha Pratim Das, Arun Kumar Majumdar

Indian Classical Dance is an over 5000 years' old multi-modal language for expressing emotions. Preservation of dance through multimedia technology is a challenging task. In this paper, we develop a system to generate a parseable representation of a dance performance. The system will help to preserve intangible heritage, annotate performances for better tutoring, and synthesize dance performances. We first attempt to capture the concepts of the basic steps of an Indian Classical Dance form, named Bharatanatyam Adavus, in an ontological model. Next, we build an event-based low-level model that relates the ontology of Adavus to the ontology of multi-modal data streams (RGB-D of Kinect in this case) for a computationally realizable framework. Finally, the ontology is used for transcription into Labanotation. We also present a transcription tool for encoding the performances of Bharatanatyam Adavus to Labanotation and test it on our recorded data set. Our primary aim is to document the complex movements of dance in terms of Labanotation using the ontology.

Tanwi Mallick, Partha Pratim Das, Arun Kumar Majumdar

Bharatanatyam, an Indian Classical Dance form, represents the rich cultural heritage of India. Analysis and recognition of such dance forms are critical for the preservation of cultural heritage. Like in most dance forms, a Bharatanatyam dancer performs in synchronization with structured rhythmic music, called Sollukattu, which comprises instrumental beats and vocalized utterances (bols) to create a rhythmic music structure. Computer analysis of Bharatanatyam, therefore, requires a structural analysis of Sollukattus. In this paper, we use speech processing techniques to recognize bols. Exploiting the predefined structures of Sollukattus and the detected bols, we recognize the Sollukattu. We estimate the tempo period by two methods. Finally, we generate a complete annotation of the audio signal by beat marking. For this, we also use the information of beats detected from the onset envelope of a Sollukattu signal. For training and test, we create a data set for Sollukattus and annotate them. We achieve 85% accuracy in bol recognition, 95% in Sollukattu recognition, 96% in tempo period estimation, and over 90% in beat marking. This is the maiden attempt to fully structurally analyze the music of an Indian Classical Dance form and the use of speech processing techniques for beat marking.

Tanwi Mallick, Prasanna Balaprakash, Eric Rask et al.

Highway traffic modeling and forecasting approaches are critical for intelligent transportation systems. Recently, deep-learning-based traffic forecasting methods have emerged as state of the art for a wide range of traffic forecasting tasks. However, these methods require a large amount of training data, which needs to be collected over a significant period of time. This can present a number of challenges for the development and deployment of data-driven learning methods for highway networks that suffer from lack of historical data. A promising approach to address this issue is transfer learning, where a model trained on one part of the highway network can be adapted for a different part of the highway network. We focus on diffusion convolutional recurrent neural network (DCRNN), a state-of-the-art graph neural network for highway network forecasting. It models the complex spatial and temporal dynamics of the highway network using a graph-based diffusion convolution operation within a recurrent neural network. DCRNN cannot perform transfer learning, however, because it learns location-specific traffic patterns, which cannot be used for unseen regions of the network. To that end, we develop a new transfer learning approach for DCRNN, where a single model trained on data-rich regions of the highway network can be used to forecast traffic on unseen regions of the highway network. We evaluate the ability of our approach to forecast the traffic on the entire California highway network with one year of time series data. We show that TL-DCRNN can learn from several regions of the California highway network and forecast the traffic on the unseen regions of the network with high accuracy. Moreover, we demonstrate that TL-DCRNN can learn from San Francisco region traffic data and can forecast traffic on the Los Angeles region and vice versa.

Tanwi Mallick, Partha Pratim Das, Arun Kumar Majumdar

Understanding the underlying semantics of performing arts like dance is a challenging task. Dance is multimedia in nature and spans over time as well as space. Capturing and analyzing the multimedia content of the dance is useful for the preservation of cultural heritage, to build video recommendation systems, to assist learners to use tutoring systems. To develop an application for dance, three aspects of dance analysis need to be addressed: 1) Segmentation of the dance video to find the representative action elements, 2) Matching or recognition of the detected action elements, and 3) Recognition of the dance sequences formed by combining a number of action elements under certain rules. This paper attempts to solve three fundamental problems of dance analysis for understanding the underlying semantics of dance forms. Our focus is on an Indian Classical Dance (ICD) form known as Bharatanatyam. As dance is driven by music, we use the music as well as motion information for key posture extraction. Next, we recognize the key postures using machine learning as well as deep learning techniques. Finally, the dance sequence is recognized using the Hidden Markov Model (HMM). We capture the multi-modal data of Bharatanatyam dance using Kinect and build an annotated data set for research in ICD.