DG
Debasis Ganguly
University of Glasgow, IBM Research - Ireland
Research areas: Computer Science · Artificial Intelligence · Information Retrieval and Search Behavior · Topic Modeling · Recommender Systems and Techniques
0000-0003-0050-7138
Employment
-
University of Glasgow Lecturer (Asst. Professor)2021 - Present
-
IBM Research - Ireland Research Staff Member2016 - 2021
Education
Education history is unavailable.
Projects & Funding
Projects & funding information is unavailable.
Publications (120)
- Generate to Accelerate: Improved Reranking via LLM-Generated Pivot Documents Save
- Predicting Partial Answer Quality and Utility in Agentic Retrieval-Augmented Generation Save
- Robustness of IR Models to Collection Growth Save
- Introduction to the Special Issue on Query Performance Prediction Save
- RAQG-QPP: Query Performance Prediction with Retrieved Query Variants and Retrieval-Augmented Query Generation Save
- A Theoretical Framework for Risk Analysis of Stochastic Rankers Save
- Relaxed Term Matching for Neural Axiomatic Diagnostics Save
- APR: Adaptive Personalised Reranking For Conversational Search Save
- Optimal Re-Ranking Depth Save
- Towards a Relevance Posterior in Neural Information Access Save
- Projection-Displacement-Based Query Performance Prediction for Embedded Space of Dense Retrievers Save
- I'm unable to view or analyze images directly. If you can describe the figure, I can help create appropriate descriptive text based on your description. SuiteEval: Simplifying Retrieval Benchmarks Save
- LURE-RAG: Lightweight Utility-Driven Reranking for Efficient RAG Save
- Predicting Retrieval Utility and Answer Quality in Retrieval-Augmented Generation Save
- Findings from the Second Shared Task on Spoken Query Cross-Lingual Information Retrieval for Indic Languages: SqCLIR at FIRE 2025 Save
- Overview of the HASOC Track at FIRE 2025: Abusive Meme Identification — Shadows Behind the Laughter Save
- Prompt Recovery for Misinformation Detection at FIRE 2025 Save
- Soft Prompt Improves Direct Search Responses Save
- HF-RAG: Hierarchical Fusion-based RAG with Multiple Sources and Rankers Save
- ProActLLM: Proactive Conversational Information Seeking with Large Language Models Save