Natwar Modani
Research Subjects
Natwar Modani has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Inderjeet Nair
Adobe Research India
2 shared publications Abhilasha Sancheti
Adobe Res, Big Data Experience Lab, Bangalore, Karnataka, India
1 shared publicationAnirudh Kanfade
NITK
1 shared publication- Anurag Maurya
Department of Botany, Shivaji College, University of Delhi, New Delhi 110027, India
1 shared publication - Aparna Garimella
Adobe Research India
1 shared publication - Arya Mazumdar
Halicioglu Data Science Institute University of California San Diego
1 shared publication
Latest Publications
- 16 June 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (7) , 149
Improving Selective Classification with Pairwise Queries for Binary Classification
- 16 June 2026
- Published by Springer Nature in Machine Learning
- Vol. 115 (7) , 149
Improving Selective Classification with Pairwise Queries for Binary Classification
- Open Access
GenSco: Can Question Decomposition based Passage Alignment improve Question Answering?
Retrieval augmented generation (RAG) with large language models (LLMs) for Question Answering (QA) entails furnishing relevant context within the prompt to facilitate the LLM in answer generation. During the generation, inaccuracies or hallucinations frequently occur due to two primary factors: inadequate or distracting context in the prompts, and the inability of LLMs to effectively reason through the facts. In this paper, we investigate whether providing aligned context via a carefully selected passage sequence leads to better answer generation by the LLM for multi-hop QA. We introduce, "GenSco", a novel approach of selecting passages based on the predicted decomposition of the multi-hop questions}. The framework consists of two distinct LLMs: (i) Generator LLM, which is used for question decomposition and final answer generation; (ii) an auxiliary open-sourced LLM, used as the scorer, to semantically guide the Generator for passage selection. The generator is invoked only once for the answer generation, resulting in a cost-effective and efficient approach. We evaluate on three broadly established multi-hop question answering datasets: 2WikiMultiHop, Adversarial HotPotQA and MuSiQue and achieve an absolute gain of $15.1$ and $5.9$ points in Exact Match score with respect to the best performing baselines over MuSiQue and 2WikiMultiHop respectively.Open AccessGenSco: Can Question Decomposition based Passage Alignment improve Question Answering?
Retrieval augmented generation (RAG) with large language models (LLMs) for Question Answering (QA) entails furnishing relevant context within the prompt to facilitate the LLM in answer generation. During the generation, inaccuracies or hallucinations frequently occur due to two primary factors: inadequate or distracting context in the prompts, and the inability of LLMs to effectively reason through the facts. In this paper, we investigate whether providing aligned context via a carefully selected passage sequence leads to better answer generation by the LLM for multi-hop QA. We introduce, "GenSco", a novel approach of selecting passages based on the predicted decomposition of the multi-hop questions}. The framework consists of two distinct LLMs: (i) Generator LLM, which is used for question decomposition and final answer generation; (ii) an auxiliary open-sourced LLM, used as the scorer, to semantically guide the Generator for passage selection. The generator is invoked only once for the answer generation, resulting in a cost-effective and efficient approach. We evaluate on three broadly established multi-hop question answering datasets: 2WikiMultiHop, Adversarial HotPotQA and MuSiQue and achieve an absolute gain of $15.1$ and $5.9$ points in Exact Match score with respect to the best performing baselines over MuSiQue and 2WikiMultiHop respectively. - Open Access
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 2516-2526
Exploiting Language Characteristics for Legal Domain-Specific Language Model Pretraining
Inderjeet Nair, Natwar Modani. Findings of the Association for Computational Linguistics: EACL 2023. 2023.
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 2516-2526
Open AccessExploiting Language Characteristics for Legal Domain-Specific Language Model Pretraining
Inderjeet Nair, Natwar Modani. Findings of the Association for Computational Linguistics: EACL 2023. 2023.
- Open Access
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 883-893
A Neural CRF-based Hierarchical Approach for Linear Text Segmentation
Inderjeet Nair, Aparna Garimella, Balaji Vasan Srinivasan, Natwar Modani, Niyati Chhaya, Srikrishna Karanam, Sumit Shekhar. Findings of the Association for Computational Linguistics: EACL 2023. 2023.
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 883-893
Open AccessA Neural CRF-based Hierarchical Approach for Linear Text Segmentation
Inderjeet Nair, Aparna Garimella, Balaji Vasan Srinivasan, Natwar Modani, Niyati Chhaya, Srikrishna Karanam, Sumit Shekhar. Findings of the Association for Computational Linguistics: EACL 2023. 2023.
- 1 January 2021
- Published by Springer Nature
- p. 405-419
Detecting Document Versions and Their Ordering in a Collection
- 1 January 2021
- Published by Springer Nature
- p. 405-419
Detecting Document Versions and Their Ordering in a Collection
- 7 October 2019
- Published by Instituto Politecnico Nacional/Centro de Investigacion en Computacion in Computación y Sistemas
- Vol. 23 (3) , 1033-1042
Understanding Blogs through the Lens of Readers’ Comments
Understanding Blogs through the Lens of Readers’ Comments
- 7 October 2019
- Published by Instituto Politecnico Nacional/Centro de Investigacion en Computacion in Computación y Sistemas
- Vol. 23 (3) , 1033-1042
Understanding Blogs through the Lens of Readers’ Comments
Understanding Blogs through the Lens of Readers’ Comments
- 25 December 2015
- Published by Springer Nature
- p. 169-184
Creating Diverse Product Review Summaries: A Graph Approach
- 25 December 2015
- Published by Springer Nature
- p. 169-184
Creating Diverse Product Review Summaries: A Graph Approach