3 papers
cs.CL2026
Are Arithmetic Heuristic Neurons Form-Invariant? A Mechanistic Analysis of Symbols, Text, and Code in LLMs
Sharath Naganna, Tanvir Ahmed Sijan, Uddipta Kalita
Large language models often succeed on one formulation of a problem while failing on an equivalent formulation. Whether these failures arise from distinct internal circuits or diff…
cs.CL2026
Beyond Clean Text: Evaluating Encoder and Decoder Robustness for Bangla Event Detection in Noisy Text
Tanvir Ahmed Sijan, S. M Golam Rifat, Nayeemul Islam +1
Event detection (ED) systems are typically evaluated on clean, curated text, leaving their robustness to real-world noise largely unexplored, particularly for low-resource language…
cs.CL2026
BanglaSocialBench: A Benchmark for Evaluating Sociopragmatic and Cultural Alignment of LLMs in Bangladeshi Social Interaction
Tanvir Ahmed Sijan, S. M Golam Rifat, Pankaj Chowdhury Partha +2
Large Language Models have demonstrated strong multilingual fluency, yet fluency alone does not guarantee socially appropriate language use. In high-context languages, communicativ…