Nihar Sanda

Nihar Sanda

Auto-generated profile, based on publicly available information

Nihar Sanda

The Institute for Experiential AI, Northeastern University, Boston, MA 02115, United States

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2026 in Bioinformatics
Verified
Journal Article
Open access
NSNihar Sanda
Nihar Sanda
BGBenjamin M Gyori
Benjamin M Gyori
VQVito Quaranta
Vito Quaranta
AGAuroop Ganguly
Auroop Ganguly
APAyan Paul
Ayan Paul
Research on biological mechanisms and disease processes is limited by fragmented findings across unstructured text in publications. Question answering and hypothesis generation that can reason across multiple sources can overcome this limitation. However, Large language models (LLMs) are prone to inaccuracies and lack clear provenance to primary evidence. Retrieval augmented generation approaches that have provenance to the original source of evidence address these shortcomings. However, the response richness is dependent on the retrieval process design. Current approaches often fail to produce responses requiring multi-hop reasoning across multiple domains. To address this, we propose eGoT, which combines automated knowledge graph construction from biomedical literature with a novel graph-of-thoughts approach to query the knowledge base and construct comprehensive responses to natural language questions. Given a corpus of documents, eGoT first uses an LLM-based pipeline to identify and normalize entities and relationships and constructs graph and vector databases. Given an input question, eGoT performs multi-round LLM-based querying of the databases to construct a response. Benchmarking on datasets like MultiHopRAG, HotpotQA, and Ultradomain demonstrates eGoT's superiority over state-of-the-art retrieval methods, including HopRAG, SireRAG, HiRAG, and HippoRAG. We demonstrate eGoT on two biomedical use cases: (i) generate responses to domain expert-curated questions on small cell lung cancer using 1046 PubMed Central publications, and (ii) demonstrate eGoT's ability to find plausible connections between Lupus and climate factors (UV exposure) that affect disease trajectory. https://github.com/NNeuralDynamics/eGOT.git.

13 April 2026
Verified
Conference Paper
Open access
MLMingyi Li
Mingyi Li
HYHuiru Yang
Huiru Yang
NSNihar Sanda
Nihar Sanda
MDMaitraye Das
Maitraye Das
Collaborative ideation tools like digital whiteboards are widely used by designers, academics, and creative practitioners; yet most ideation tools are inaccessible to blind or low vision (BLV) users. Informed by prior work on whiteboarding challenges encountered by BLV users and our formative study with eight sighted whiteboard users, we built Idea11y, a whiteboard plug-in that provides a hierarchical, editable text outline of board content, augmented with audio cues and voice coding. Findings from evaluation with thirteen BLV screen reader users revealed how Idea11y supported BLV users’ understanding of clustering structure and streamlined their process to author, synthesize, and prioritize ideas on the board. Collaborative ideation sessions with six BLV-sighted dyads demonstrated how BLV users used Idea11y to develop collaboration awareness and coordinate actions with sighted collaborators. Drawing on this, we discuss ways to move beyond implicit visual norms in established ideation frameworks and practical considerations for future accessible ideation systems.