7 papers
BODHI: Do LLMs Branch Out and Discover Heterogeneous Inferences?
Soumadeep Saha, Krish Sharma, Akshay Chaturvedi +1
Although reinforcement learning with verifiable rewards (RLVR) has improved the performance of large language models (LLMs) across a variety of reasoning tasks, there is significan…
TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability
Krish Sharma, Omar Naim, Soumadeep Saha +3
Recent work has promoted task-aware layer pruning as a way to improve model performance on particular tasks, as shown by TALE. In this paper, we investigate when such improvements…
KisMATH: Do LLMs Have Knowledge of Implicit Structures in Mathematical Reasoning?
Soumadeep Saha, Akshay Chaturvedi, Saptarshi Saha +2
Chain-of-thought (CoT) traces have been shown to improve performance of large language models on a plethora of reasoning tasks, yet there is no consensus on the mechanism by which…
sudoLLM: On Multi-role Alignment of Language Models
Soumadeep Saha, Akshay Chaturvedi, Joy Mahapatra +1
User authorization-based access privileges are a key feature in many safety-critical systems, but have not been extensively studied in the large language model (LLM) realm. In this…
On Measuring Intrinsic Causal Attributions in Deep Neural Networks
Saptarshi Saha, Dhruv Vansraj Rathore, Soumadeep Saha +2
Quantifying the causal influence of input features within neural networks has become a topic of increasing interest. Existing approaches typically assess direct, indirect, and tota…
Language Models are Crossword Solvers
Soumadeep Saha, Sutanoya Chakraborty, Saptarshi Saha +1
Crosswords are a form of word puzzle that require a solver to demonstrate a high degree of proficiency in natural language understanding, wordplay, reasoning, and world knowledge,…