This study investigates whether a custom-built model that only pulls results from a verified domain can provide more accurate literature analysis than a standard generative language model such as ChatGPT, and suggests that a retrieval-constrained system may improve accuracy and traceability relative to a generative baseline.
The growing volume of scientific literature presents a significant challenge for researchers to identify relevant methods, compare findings, and build upon existing work. Although advances in artificial intelligence have led to new tools for literature analysis, many of these tools rely on generative models whose outpu...
K. Dey· UC Riverside Undergraduate R...· 0 citations
What makes great scientists great? Even as AI systems start to make progress on open problems, scientists remain far ahead of them at sensing which prior idea, buried in an ever-growing archive of research, a new problem needs. To study this skill, we draw on researchers who know firsthand which earlier work advanced t...
Sohyeon Kim, Yoonho Lee, Bo Liu et al.· 0 citations
Over the past years, large language models (LLMs) have gained significant attention and have led to major breakthroughs in natural language processing (NLP). Despite their remarkable efficiency and contextual understanding, LLMs still remain limited by their tendency to generate hallucinated text, their reliance on sta...
Vasileios P. Rekkas, Nicholas Vretos· Technologies· 0 citations
A domain-expert evaluation of faithfulness for AquiLLM, an open-weight, offline RAG-LLM platform designed to support scientific research groups in the use and preservation of tacit and formal knowledge, is presented.
Bernie Boscoe, S. Saikrishnan, Vikram Seenivasan et al.· arXiv.org· 1 citation
Scientific research increasingly depends on curating experimental data from literature. Scientific data curation requires verifying exact experimental conditions (e.g., specific assays, clinical outcomes, and data availability) across long, heterogeneous files, which is a highly time-consuming process. Existing benchma...
SnapJudge is trained: an image-based detector operating on screenshots of both prose and proofs from papers, substantially outperforming both Pangram 4 and the authors' text-based detector.