SocialNLP Research Lab
School of Computing, Macquarie University, Sydney, Australia — part of the Macquarie NLP Lab

Last updated: 19 Nov 2025

SocialNLP Research Lab

Highlights

Recent achievements and recognitions from our group.

Nov 2025 — 🎉 Congrats to Juan (Ada) Ren for her paper acceptance at IEEE Big Data 2025!
Nov 2025 — 🌟 Congrats to Usman Naseem for receiving the Outstanding Senior Area Chair Award at EMNLP 2025!
Oct 2025 — 🎉 Congrats to Yiran Zhang for his AAAI 2026 Demo Track paper, and to Usman Naseem for his Main Track paper LLM2CLIP.
Sep 2025 — 🎉 Congrats to Afrozah Nadeem for winning Best Paper and Best Presentation at CASE @ RANLP 2025.
Aug 2025 — 🏆 10 papers (3 Main, 4 Findings, 2 Industry Track, 1 Workshop) accepted at EMNLP 2025. Congrats to the team!
Aug 2025 — 🎓 Two invited tutorials on AI Alignment with Human Values and Preferences to be delivered at ALTA 2025 and AJCAI 2025.

Research Vision

At SocialNLP, we explore how language models can understand, reason, and act in ways that align with human values and social contexts — advancing research at the intersection of AI Alignment, Social Good, and Online Trust & Safety.

Research Areas

AI Alignment

Advancing methods to align LLMs with human values, and preferences.

Value Alignment — Ensuring models act in ways consistent with human ethics, integrity, and the HHH principles (Helpful, Harmless, Honest). Also, reflecting diverse cultural and moral perspectives beyond a single worldview.
Safety Alignment — Developing defenses against jailbreaks, adversarial attacks, and unsafe content generation.
Reasoning Alignment — Promoting logical consistency, factual grounding, and transparent reasoning chains to minimize hallucinations.
Cultural Alignment — Embedding diverse moral and social perspectives to ensure global inclusivity and fairness in LLM behavior.

NLP for Social Good

Applying NLP technologies to enhance societal welfare, inclusivity, and equitable access to information across diverse communities.

Healthcare NLP — Combating misinformation and supporting public health communication through evidence-grounded LLMs.
Low-Resource Languages — Developing inclusive datasets and language tools for under-represented regions.
Climate & Crisis Communication — Enabling real-time, multilingual information flow during emergencies and disasters.

Online Trust and Safety

Building transparent, interpretable, and ethically aware AI systems that preserve trust and mitigate online harms at scale.

Misinformation Detection — Identifying and explaining false or AI-generated content across text and media.
Toxicity & Abuse Prevention — Detecting hate speech, harassment, and coordinated manipulation campaigns.
Content Moderation — Aligning automated filters with community guidelines and global cultural norms.
Trustworthy Agents — Designing accountable conversational models with auditability and user-level transparency.

People

Faculty Lead

Faculty Lead

Dr. Usman Naseem

Macquarie University, Australia

HDR @ SocialNLP

PhD Student

Gautam Siddharth Kashyap

Macquarie University, Australia

PhD Student

Kaixuan Ren (Victor)

Macquarie University, Australia

MRes Student

Utsav Maskey

Macquarie University, Australia

MRes Student

Afrozah Nadeem

Macquarie University, Australia

MRes Student

Juan (Ada) Ren

Macquarie University, Australia

MRes Student

Yiran Zhang (Grant)

Macquarie University, Australia

RA @ SocialNLP

Ebad Shabbir (Remote)
Rafiq Ali (Remote)
Abdullah Mohammad (Remote)
Sushant Kumar Ray (Remote)
Anudeex Shetty (Remote)
Zhihao Zhang (Remote)
Mo Wang (Remote)
Syed Talal Ahmad (Past)
Kaixuan Ren (Past --> Current PhD Student @ SocialNLP)

Affiliated with @ SocialNLP

Lingfeng Zhong — Co-Advisor @ Macquarie University, Sydney, Australia
Abid Ali — Co-Advisor @ Macquarie University, Sydney, Australia
Saleh Afzoon — Co-Advisor @ Macquarie University, Sydney, Australia
Pranav Bhandari — Co-Advisor @ University of Western Australia, Perth, Australia
Ahmed Abbasi — Co-Advisor @ James Cook University, Queensland, Australia
Muhammad Islam — Co-Advisor @ James Cook University, Queensland, Australia
Shuchang Ye — Co-Advisor @ at University of Sydney, Sydney, Australia
Yongpei Ma — Co-Advisor @ University of Sydney, Sydney, Australia
Pengyu Wang — Co-Advisor @ University of Sydney, Sydney, Australia
Yupeng Zhang — Co-Advisor @ University of Sydney, Sydney, Australia
Qixuan Hu — Co-Advisor @ University of Sydney, Sydney, Australia
Xiao Wang — Co-Advisor @ University of Sydney, Sydney, Australia

Recent Publications

EMNLP 2025

EMNLP Main 2025
Kashyap, Gautam Siddharth, Mark Dras, and Usman Naseem. “Too Helpful, Too Harmless, Too Honest or Just Right?” arXiv preprint arXiv:2509.08486 (2025).
EMNLP Main 2025
Zhong, Jiayou, Anudeex Shetty, Chao Jia, Xuanrui Lin, and Usman Naseem. “Pluralistic Alignment for Healthcare: A Role-Driven Framework.” arXiv preprint arXiv:2509.10685 (2025).
EMNLP Main 2025
Bhandari, Pranav, Nicolas Fay, Michael Wise, Amitava Datta, Stephanie Meek, Usman Naseem, and Mehwish Nasim. “Can LLM Agents Maintain a Persona in Discourse?” arXiv preprint arXiv:2502.11843 (2025).
EMNLP Findings 2025
Zhang, Zhihao, Yiran Zhang, Xiyue Zhou, Liting Huang, Imran Razzak, Preslav Nakov, and Usman Naseem. “From Generation to Detection: A Multimodal Multi-Task Dataset for Benchmarking Health Misinformation.” arXiv preprint arXiv:2505.18685 (2025).
EMNLP Findings 2025
Chandna, Bhavik, Mariam Aboujenane, and Usman Naseem. “ExtremeAIGC: Benchmarking LMM Vulnerability to AI-Generated Extremist Content.” arXiv preprint arXiv:2503.09964 (2025).
EMNLP Findings 2025
Maskey, Utsav, Chencheng Zhu, and Usman Naseem. “Benchmarking Large Language Models for Cryptanalysis and Mismatched-Generalization.” arXiv preprint arXiv:2505.24621 (2025).
EMNLP Findings 2025
Zhang, Yiran, Mo Wang, Xiaoyang Li, Kaixuan Ren, Chencheng Zhu, and Usman Naseem. “TurnBench-MS: A Benchmark for Evaluating Multi-Turn, Multi-Step Reasoning in Large Language Models.” arXiv preprint arXiv:2506.01341 (2025).
EMNLP Industry Track 2025
Azeez, Mohammad Anas, Rafiq Ali, Ebad Shabbir, Zohaib Hasan Siddiqui, Gautam Siddharth Kashyap, Jiechao Gao, and Usman Naseem. “Truth, Trust, and Trouble: Medical AI on the Edge.” (2025).
EMNLP Industry Track 2025
Siddiqui, Zohaib Hasan, Jiechao Gao, Ebad Shabbir, Mohammad Anas Azeez, Rafiq Ali, Gautam Siddharth Kashyap, and Usman Naseem. “LLMs on a Budget? Say HOLA.” (2025).

CASE @ RANLP 2025

CASE@RANLP 2025 (Best Paper Award)
Nadeem, Afrozah, Mark Dras, and Usman Naseem. “Steering Towards Fairness: Mitigating Political Bias in LLMs.” arXiv preprint arXiv:2508.08846 (2025).

ACL 2025

ACL Main 2025
Shetty, Anudeex, Amin Beheshti, Mark Dras, and Usman Naseem. “VITAL: A New Dataset for Benchmarking Pluralistic Alignment in Healthcare.” arXiv preprint arXiv:2502.13775 (2025).
ACL Main 2025
Yuan, Jiahao, Zixiang Di, Zhiqing Cui, Guisong Yang, and Usman Naseem. “ReflectDiffu: Reflect between Emotion-intent Contagion and Mimicry for Empathetic Response Generation via a RL-Diffusion Framework.” arXiv preprint arXiv:2409.10289 (2024).
ACL Main 2025
Yuan, Jiahao, Dehui Du, Hao Zhang, Zixiang Di, and Usman Naseem. “Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up.” arXiv preprint arXiv:2410.12323 (2024).

Contact

Faculty Lead

Dr. Usman Naseem

Macquarie University, Australia

📧 usman.naseem@mq.edu.au

School of Computing, Macquarie University, Sydney, NSW 2113, Australia