CV

Postdoctoral Research Associate

University of Edinburgh, Scotland

Developed neural IR models and led projects involving Knowledge Graph semantics and LLMs.

• Developed a novel neural IR model that integrates Knowledge Graph semantics to learn query-specific document representations, outperforming several hard baselines by 40-60%.
• Led the development of the Generative Relevance Feedback method leveraging LLMs to derive query expansion terms, improving precision by 5-19%.
• Collaborated on mitigating LLM hallucination tendencies, improving recall by 2-4%.
• Developed personalized conversational systems in collaboration with Radboud University.

Postdoctoral Research Associate

University of Glasgow, Scotland

Focused on integrating Knowledge Graph semantics with IR systems and entity-oriented search.

• Collaborated on integrating sparse retrieval models with Knowledge Graph embeddings.
• Worked on large-scale dialogue datasets for multi-session conversational agents.

Postdoctoral Research Fellow

University of New Hampshire, Durham, USA

Conducted advanced research in entity-oriented search and information retrieval systems.

• Proposed novel methods for learning query-specific latent entity spaces for IR, outperforming hard baselines by 40-60%.
• Integrated Wikipedia-based entity embeddings into entity ranking models, achieving a 54% improvement on large-scale datasets.
• Collaborated on BERT-based entity representation models, published at IJCKG 2023.
• Focused on using entity retrieval techniques to improve document relevance.

Graduate Research Assistant

University of New Hampshire, Durham, USA

Focused on entity-specific embeddings and neural information retrieval systems.

• Developed entity-specific embedding models published at SIGIR and CIKM.
• Demonstrated improvements in entity ranking by 13-42% on large-scale datasets.
• Proposed a novel probabilistic model for passage ranking, achieving state-of-the-art results.

PhD

University of New Hampshire

Computer Science

Master of Science

University of New Hampshire

Computer Science

Master of Science

University of Calcutta

Computer Science

Bachelor of Science

University of Calcutta

Computer Science

TREC iKAT: A Test Collection for Evaluating Conversational and Interactive Knowledge Assistants

Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

Doing Personal LAPS: LLM-Augmented Dialogue Construction for Personalized Multi-Session Conversational Search

Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

Adaptive Latent Entity Expansion for Document Retrieval

Proceedings of the First Knowledge-Enhanced Information Retrieval Workshop, ECIR

DREQ: Document Re-Ranking Using Entity-based Query Understanding

Proceedings of the 46th European Conference on Information Retrieval (ECIR)

DyVo: Dynamic Vocabularies for Learned Sparse Retrieval with Entities

Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Generative and Pseudo-Relevant Feedback for Sparse, Dense, and Learned Sparse Retrieval

Proceedings of the Workshop on Large Language Models’ Interpretation and Trustworthiness (CIKM)

Generative Relevance Feedback with Large Language Models

Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

Neural Entity Context Models

Proceedings of the 12th International Joint Conference on Knowledge Graphs (IJCKG)

An Entity-Oriented Approach for Answering Topical Information Needs

Proceedings of the 44th European Conference on Information Retrieval (ECIR)

Wikimarks: Harvesting Relevance Benchmarks from Wikipedia

Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

BERT-ER: Query-Specific BERT Entity Representations for Entity Ranking

Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

Predicting Guiding Entities for Entity Aspect Linking

Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM)

Entity Retrieval Using Fine-Grained Entity Aspects

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

Why Does This Entity Matter? Support Passage Retrieval for Entity Retrieval

Proceedings of the 2019 ACM SIGIR International Conference on Theory of Information Retrieval (ICTIR)

Best Paper Honourable Mention

ECIR

Outstanding Reviewer Award

ECIR

Distinguished Reviewer Award

CIKM

Outstanding Reviewer Award

ECIR

Dissertation Year Fellowship

University of New Hampshire

ACM Travel Grants

ACM

INSPIRE Fellowship

Government of India