Anjishnu Mukherjee

Pronunciation: /ʌnˈdʒɪʃnuː/ (un-JISH-noo)

I am a fourth-year Ph.D. candidate in Computer Science at George Mason University, advised by Antonis Anastasopoulos and Ziwei Zhu in the GMU NLP Lab.

Research directions

My research studies how language and multimodal systems represent cultures and the people who constitute them through three connected themes.

Measurement: designing evaluations for socio-cultural bias, multilingual behavior, and cultural grounding across languages, regions, and modalities (Global Voices, South Asian biases, BiasDora, Tower of Babel).
Modeling: studying how data, metadata, representations, and model mechanisms shape cultural and linguistic behavior (Metadata conditioning, KnowBias, Global Gallery).
Intervention: developing methods to localize, debias, and adapt AI systems so they respond appropriately to cultural and regional context (Crossroads, Breaking Bias, Metadata conditioning).

I am currently on the 2027 job market for academic roles, including postdoctoral and tenure-track positions.

News

May 2026: Will be joining my third Applied Science internship at Amazon.
May 2026: Jinhao's paper on "Knowing Bias, Doing Better" was accepted to ICML 2026 as a poster.
May 2026: Rinki's paper on measuring South Asian biases was accepted to TrustNLP at ACL 2026.
Apr 2026: Passed my Ph.D. proposal defense and advanced to candidacy.
Dec 2025: Completed my M.S. in Computer Science with an Outstanding Academic Achievement award.

Selected Publications

Selected peer-reviewed and under-review research. For the full list, see Google Scholar.

JavaScript is disabled, so this is a compact fallback list. Enable JavaScript for filters, figures, and abstracts.

Knowing Bias, Doing Better: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement ICML '26 (Poster), 2026. Jinhao Pan, Chahat Raj, Anjishnu Mukherjee, Sina Mansouri, Bowen Wei, Shloka Yada, Ziwei Zhu.
Lost in the Tower of Babel: The Adverse Effects of Incidental Multilingualism in LLMs ArXiv, 2026. Anjishnu Mukherjee, Chutong Meng, Antonios Anastasopoulos.
Metadata Conditioned Large Language Models for Localization ArXiv, 2026. Anjishnu Mukherjee, Ziwei Zhu, Antonios Anastasopoulos.
Measuring South Asian Biases in Large Language Models TrustNLP @ ACL 2026 (Poster), 2025. Mamnuya Rinki, Chahat Raj, Anjishnu Mukherjee, Ziwei Zhu.
Crossroads of Continents: Automated Artifact Extraction for Cultural Adaptation with Large Multimodal Models WACV '25 (Oral and Poster), 2025. Anjishnu Mukherjee, Ziwei Zhu, Antonios Anastasopoulos.
BiasDora: Exploring Hidden Biased Associations in Vision-Language Models EMNLP Findings '24 (Poster), 2024. Chahat Raj, Anjishnu Mukherjee, Aylin Caliskan, Antonios Anastasopoulos, Ziwei Zhu.
Breaking Bias, Building Bridges: Evaluation and Mitigation of Social Biases in LLMs via Contact Hypothesis AIES '24 (Oral), 2024. Chahat Raj*, Anjishnu Mukherjee*, Aylin Caliskan, Antonios Anastasopoulos, Ziwei Zhu.
Global Gallery: The Fine Art of Painting Culture Portraits through Multilingual Instruction Tuning NAACL '24 (Poster), 2024. Anjishnu Mukherjee, Aylin Caliskan, Ziwei Zhu, Antonios Anastasopoulos.
Global Voices, Local Biases: Socio-cultural Prejudices across Languages EMNLP '23 (Poster), 2023. Anjishnu Mukherjee*, Chahat Raj*, Ziwei Zhu, Antonios Anastasopoulos.