Papers by Mohna Chakraborty
Is Large Language Model Performance on Reasoning Tasks Impacted by Different Ways Questions Are Asked? (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on Large Language Models (LLMs) have not investigated the impact of question types on LLM performance. |
| Approach: | They evaluate the performance of five Large Language Models on reasoning tasks . they use quantitative reasoning tasks and deductive reasoning tasks to evaluate the models . |
| Outcome: | The results show that Reasoning accuracy does not correlate with final selection accuracy. |
Structured Moral Reasoning in Language Models: A Value-Grounded Evaluation Framework (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) are increasingly deployed in domains requiring moral understanding, yet their reasoning often remains shallow and misaligned with human reasoning. |
| Approach: | They propose a value-grounded framework for evaluating and distilling structured moral reasoning in large language models. |
| Outcome: | The proposed framework evaluates 12 open-source models across four moral datasets. |
Zero-shot Approach to Overcome Perturbation Sensitivity of Prompts (2023.acl-long)
Copied to clipboard
| Challenge: | Recent studies have demonstrated that natural-language prompts can help to leverage the knowledge learned by pre-trained language models for the binary sentence-level sentiment classification task. |
| Approach: | They propose to use few-shot learning settings to fine-tune the sentiment classification model using manual or automatically generated prompts. |
| Outcome: | The proposed method outperforms the base prompt and the prompts generated using few-shot learning for the binary sentence-level sentiment classification task. |