Papers by Bhaskar Ramasubramanian

4 papers
Small Models Struggle to Learn from Strong Reasoners (2025.findings-acl)

Copied to clipboard

Challenge: a small learning gap exists between large and small language models . long CoT data and large model responses are not beneficial for small models - a problem that may be due to the small student model's ability to handle distribution shifts.
Approach: They propose a mix distillation strategy that balances reasoning complexity by combining long and short CoT examples or reasoning from both larger and smaller models.
Outcome: The proposed strategy outperforms training on large and small models on short CoT and small model CoT.
Temporal Sampling for Forgotten Reasoning in LLMs (2026.acl-long)

Copied to clipboard

Challenge: a new metric measures the percentage of questions that were answered incorrectly during fine-tuning .
Approach: They propose a decoding strategy that draws outputs from multiple checkpoints along the training trajectory.
Outcome: The proposed method improves reasoning performance and consistency across benchmarks.
CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models (2024.emnlp-main)

Copied to clipboard

Challenge: Generative large language models (LLMs) have remarkable performance in generation tasks, but datasets used to train or fine-tune these models are often not disclosed to users.
Approach: They develop an inference time defense called CleanGen to mitigate backdoor attacks for generation tasks in large language models.
Outcome: The proposed inference time defense achieves lower attack success rates (ASR) compared to baseline defenses for all five backdoor attacks.
EDC: Effective and Efficient Dialog Comprehension For Dialog State Tracking (2024.naacl-long)

Copied to clipboard

Challenge: Existing methods for dialog state tracking face trade-offs between accuracy and efficiency . effective and efficient dialog comprehension (EDC) predicts domains, slot names and slot values of dialog state step-by-step for better accuracy .
Approach: They propose an alternative method that leverages the tree structure of the dialog state.
Outcome: The proposed approach achieves state-of-the-art JGA accuracy and is more efficient than previous models.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations