Papers by Srikanth G. Tamilselvam
ODASim: Ordered, Distinctive and Absolute Semantic Similarity for Code Explanation Evaluation (2026.findings-acl)
Copied to clipboard
Prince Kumar, Vitobha Munigala, Jaydeep Sen, Ashish Mittal, Vishwajeet Kumar, Srikanth G. Tamilselvam
| Challenge: | Existing methods for code explanations fail to distinguish correct from partially or fully incorrect explanations and their similarity scores are poorly calibrated. |
| Approach: | They propose a model-agnostic graded fine-tuning framework that learns calibrated similarity representations between code and explanations to support fine-grained supervision and evaluation. |
| Outcome: | The proposed framework improves F1 score and ECE scores on two embedding models and reduces expected calibration error. |
ConCodeEval: Evaluating Large Language Models for Code Constraints in Domain-Specific Languages (2025.acl-industry)
Copied to clipboard
Mehant Kammakomati, Sameer Pimparkhede, Srikanth G. Tamilselvam, Prince Kumar, Pushpak Bhattacharyya
| Challenge: | Large Language Models (LLMs) have demonstrated potential in code generation and natural language understanding, but they struggle with code constraints. |
| Approach: | They propose to use Large Language Models to handle constraints represented in code . they use JSON, YAML, XML, Python, and natural language to test their effectiveness . |
| Outcome: | The proposed benchmark shows that LLMs can handle code constraints better than natural language . the results suggest that conscious choice of representations can lead to optimal use of LLM in enterprise use cases involving code constraints. |
ETF: An Entity Tracing Framework for Hallucination Detection in Code Summaries (2025.acl-long)
Copied to clipboard
Kishan Maharaj, Vitobha Munigala, Srikanth G. Tamilselvam, Prince Kumar, Sayandeep Sen, Palani Kodeswaran, Abhijit Mishra, Pushpak Bhattacharyya
| Challenge: | Recent advances in large language models have significantly enhanced their ability to understand both natural language and code, but are prone to hallucinations. |
| Approach: | They propose a first-of-its-kind dataset, CodeSumEval, with 10K samples, curated specifically for hallucination detection in code summarisation. |
| Outcome: | The proposed framework has a 73% F1 score and is curated specifically for detection of hallucinations in code summarisation. |