arXiv subjects

Sujit Kumar

Publications and source records attributed to Sujit Kumar.

3 recordsLinked to original sources

arXiv2026

TRIDENT and SHAKTI: A Framework and Benchmark for Hindi AI-Text Detection

As Large Language Models (LLMs) become increasingly multilingual, distinguishing human-written from AI-generated text is essential for maintaining information integrity. However, research remains English-centric, with limited benchmarks for low-resource languages such as Hindi. To address this gap, we introduce SHAKTI, a large-scale Hindi benchmark covering four genres, eight LLM families, multiple prompting strategies, and adversarial transformations. We evaluate its quality through human annotation and analyze linguistic, grammatical, and sentiment-based differences between human- and AI-generated texts. Building on these findings, we propose TRIDENT, a round-trip translation-based detection framework with three variants: TRIDENT-S (static threshold), TRIDENT-D (dynamic threshold), and TRIDENT-C (logistic regression using multi-distance RTT features). TRIDENT leverages semantic consistency between original and round-trip translated texts to identify AI-generated content. Extensive experiments show that TRIDENT outperforms conventional and supervised baselines, with TRIDENT-S achieving the best overall performance. Further evaluations across seven languages, adversarial settings, and unseen LLMs demonstrate its multilingual applicability and robustness, while highlighting sensitivity to the translation model.

cs.AI↗

arXiv2025

SciClaimHunt: A Large Dataset for Evidence-based Scientific Claim Verification

Verifying scientific claims presents a significantly greater challenge than verifying political or news-related claims. Unlike the relatively broad audience for political claims, the users of scientific claim verification systems can vary widely, ranging from researchers testing specific hypotheses to everyday users seeking information on a medication. Additionally, the evidence for scientific claims is often highly complex, involving technical terminology and intricate domain-specific concepts that require specialized models for accurate verification. Despite considerable interest from the research community, there is a noticeable lack of large-scale scientific claim verification datasets to benchmark and train effective models. To bridge this gap, we introduce two large-scale datasets, SciClaimHunt and SciClaimHunt_Num, derived from scientific research papers. We propose several baseline models tailored for scientific claim verification to assess the effectiveness of these datasets. Additionally, we evaluate models trained on SciClaimHunt and SciClaimHunt_Num against existing scientific claim verification datasets to gauge their quality and reliability. Furthermore, we conduct human evaluations of the claims in proposed datasets and perform error analysis to assess the effectiveness of the proposed baseline models. Our findings indicate that SciClaimHunt and SciClaimHunt_Num serve as highly reliable resources for training models in scientific claim verification.

cs.CL↗

arXiv2022

Multiple low-energy excitons and optical response of $d^0$ double perovskite Ba$_2$ScTaO$_6$

Large bandgap insulators are considered promising for applications such as photocatalysts, dielectric resonators and interference filters. Based on synchrotron X-ray diffraction, diffuse reflectance measurement and density functional theory, we report the crystal structure, optical response, and electronic properties of the synthesized $d^0$ double perovskite Ba$_2$ScTaO$_6$. In contrast to earlier prediction, the electronic bandgap is found to be large, $\sim 4.66$ eV. The optical response is characterized by the presence of multiple exciton modes extending up to the visible range. A detailed investigation of the direct gap excitons based on the Elliot formula is presented. Density functional theory based investigation of the electronic properties within generalized gradient approximation severely underestimates the electronic gap. To reach a quantitative agreement, we consider different available flavors of the modified-Becke-Johnson exchange-correlation potential and discuss their effects on the electronic and optical properties.

cond-mat.mtrl-sci↗