Papers by Nikhil Sharma
Neural Conversational QA: Learning to Reason vs Exploiting Patterns (2020.emnlp-main)
Copied to clipboard
| Challenge: | Neural Conversational QA tasks such as ShARC require systems to answer questions based on the contents of a given passage. |
| Approach: | They propose to modify a data-set with fewer spurious patterns to exploit them . they also propose to build a heuristic-based program to exploit spurious clues . |
| Outcome: | The proposed program exploits spurious patterns in the ShARC dataset, compared to neural models. |
Laying Anchors: Semantically Priming Numerals in Language Modeling (2024.findings-naacl)
Copied to clipboard
| Challenge: | Numeracy is the comprehension of numbers, and numerals are important for comprehension. |
| Approach: | They propose methods to semantically prime numerals by generating anchors governed by the distribution of numeral in any corpus. |
| Outcome: | The proposed methods perform better on numeracy tasks for both in-domain and out-domain numerals. |
Ranking LLM-Generated Loop Invariants for Program Verification (2023.findings-emnlp)
Copied to clipboard
Saikat Chakraborty, Shuvendu Lahiri, Sarah Fakhoury, Akash Lal, Madanlal Musuvathi, Aseem Rastogi, Aditya Senthilnathan, Rahul Sharma, Nikhil Swamy
| Challenge: | Large Language Models (LLMs) are capable of synthesizing inductive loop invariants for a class of programs in a 0-shot setting, yet require several samples to generate the correct invariant. |
| Approach: | They propose a re-ranking approach to generate inductive loop invariants using Large Language Models . they propose reranking rankers that can distinguish between correct and incorrect attempts . |
| Outcome: | The proposed method reduces the number of calls to a verifier by comparing the generated results with the original model. |
Faux Polyglot: A Study on Information Disparity in Multilingual Large Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent surge in multilingual large language models (LLMs) and Retrieval Augmented Generation (RAG) has significantly expanded conversational search across varied linguistic and cultural demographics. |
| Approach: | They found that LLMs displayed systemic bias towards information in the same language as query language in document retrieval and answer generation. |
| Outcome: | The results highlight the linguistic divide within multilingual LLMs in information search systems. |
Learning Non-linguistic Skills without Sacrificing Linguistic Proficiency (2023.acl-long)
Copied to clipboard
| Challenge: | Numeracy is the most prevalent form of non-linguistic information embedded in textual corpora. |
| Approach: | They propose a framework for non-linguistic skill injection for LLMs that incorporates information-theoretic interventions and skill-specific losses to enable the learning of strict arithmetic reasoning. |
| Outcome: | The proposed model outperforms the state-of-the-art on injected non-linguistic skills and on linguistic knowledge retention with a fraction of the non-language training data (1/4) and zero additional synthetic linguistic training data. |
Distantly Supervised Transformers For E-Commerce Product QA (2021.naacl-main)
Copied to clipboard
| Challenge: | e-commerce services often provide an instant QA system on product pages . however, user queries and CQA pairs differ significantly in language characteristics . |
| Approach: | They propose a transformer-based instant question answering system on product pages . for each user query, relevant community question answer (CQA) pairs are retrieved . their framework is able to scale to large e-commerce QA traffic . |
| Outcome: | The proposed model outperforms syntactic and semantic baselines on user queries and training with CQA pairs. |