Shahriar Kabir Nahin — accepted AI-conference papers

2026

Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models

Shahriar Kabir Nahin, Hadi Askari, Muhao Chen, Anshuman Chhabra

ICML 2026poster

Test-Time Scaling (TTS) improves LLM reasoning by exploring multiple candidate responses and then operating over this set to find the best output. A tacit premise behind TTS is that sufficiently diverse candidate pools enhance reliability. In this work, we show that this assumption in TTS introduces…

Cited by 0 Source ↗Scholar ↗

2025

TituLLMs: A Family of Bangla LLMs with Comprehensive Benchmarking

Shahriar Kabir Nahin, Rabindra Nath Nandi, Sagor Sarker, Quazi Sarwar Muhtaseem, Md Kowsher, Apu Chandraw Shill +4

ACL 2025finding

In this paper, we present TituLLMs, the first large pretrained Bangla LLMs, available in 1b and 3b parameter sizes. Due to computational constraints during both training and inference, we focused on smaller models. To train TituLLMs, we collected a pretraining dataset of approximately ∼ 37 billion t…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗