T. M. Murali

Sha Li, Ayush Sadekar, Nathan Self et al.

Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the realities of scientific literature: long-context documents, multi-modal content, and reconciling varied and inconsistent fine-grained information across multiple publications into standardized formats. These challenges are further compounded when the desired data schema or extraction ontology changes rapidly, making it difficult to re-architect or fine-tune existing systems. We present SciEx, a modular and composable framework that decouples key components including PDF parsing, multi-modal retrieval, extraction, and aggregation. This design streamlines on-demand data extraction while enabling extensibility and flexible integration of new models, prompting strategies, and reasoning mechanisms. We evaluate SciEx on datasets spanning three scientific topics for its ability to extract fine-grained information accurately and consistently. Our findings provide practical insights into both the strengths and limitations of current LLM-based pipelines.

Blessy Antony, Amartya Dutta, Sneha Aggarwal et al.

The lack of high-quality ground truth datasets to train machine learning (ML) models impedes the potential of artificial intelligence (AI) for science research. Scientific information extraction (SIE) from the literature using LLMs is emerging as a powerful approach to automate the creation of these datasets. However, existing LLM-based approaches and benchmarking studies for SIE focus on broad topics such as biomedicine and chemistry, are limited to choice-based tasks, and focus on extracting information from short and well-formatted text. The potential of SIE methods in complex, open-ended tasks is considerably under-explored. In this study, we used a domain that has been virtually ignored in SIE, namely virology, to address these research gaps. We design a unique, open-ended SIE task of extracting mutations in a given virus that modify its interaction with the host. We develop a new, multi-step retrieval augmented generation (RAG) framework called VILLA for SIE. In parallel, we curate a novel dataset of 629 mutations in ten influenza A virus proteins obtained from 239 scientific publications to serve as ground truth for the mutation extraction task. Finally, we demonstrate VILLA's superior performance using a novel and comprehensive evaluation and comparison with vanilla RAG and other state-of-the art RAG- and agent-based tools for SIE.

Aditya Bharadwaj, David Gwizdala, Yoonjin Kim et al.

Modern experiments in many disciplines generate large quantities of network (graph) data. Researchers require aesthetic layouts of these networks that clearly convey the domain knowledge and meaning. However, the problem remains challenging due to multiple conflicting aesthetic criteria and complex domain-specific constraints. In this paper, we present a strategy for generating visualizations that can help network biologists understand the protein interactions that underlie processes that take place in the cell. Specifically, we have developed Flud, an online game with a purpose (GWAP) that allows humans with no expertise to design biologically meaningful graph layouts with the help of algorithmically generated suggestions. Further, we propose a novel hybrid approach for graph layout wherein crowdworkers and a simulated annealing algorithm build on each other's progress. To showcase the effectiveness of Flud, we recruited crowd workers on Amazon Mechanical Turk to lay out complex networks that represent signaling pathways. Our results show that the proposed hybrid approach outperforms state-of-the-art techniques for graphs with a large number of feedback loops. We also found that the algorithmically generated suggestions guided the players when they are stuck and helped them improve their score. Finally, we discuss broader implications for mixed-initiative interactions in human computation games.