Reza Haffiri- Main

Positions

Director of Engagement, Department of Data Sciece and AI, Monash University, 2024-now
Head of the Language and Speech Processing, OpenStream AI, 2022-now.
ARC Future Fellow and Professor, Department of Data Sciece and AI, Monash University, 2022-now
Director of the Vision and Language Group, Department of Data Sciece and AI, Monash University, 2020-2024.

Prospective Students

For all prospective students (undergrad/msc/phd), please read this before you contact me. Due to the large volume of emails, I may not be able to answer all emails; apologies!
To international undergrad students: Please email me only if you have your own funding for internship as I do not have any funding to support interns.

Research Areas

Neuro-Symbolic AI Agents, e.g. integrating LLMs and symbolic methods for multi-modal agents that operate/learn by perception, planning, acting, and feeback
Machine Learning, e.g. deep generative models, Bayesian learning and inference, deep causal models, explainable AI, deep reinforcement learning
Knowledge Graphs and LLMs, e.g. effective approaches for hybrid LLM-KG reasoning, certifying LLM knwoledge/reasoning
Efficient and effective LLMs, e.g. better architectures, better decoding/inference, better training algorithms (long-context LLMs, KV-caching)
Continual Learning, e.g. LLM knwoledge editing, retreival augmented generation (RAG)
Reliable and Safe LLMs, e.g. alignment algorithms (RLHF), social-cultural norms, jailbreaking and defence mechanisms
Multilingual and Multimodal Foundation Models, e.g. covering low-resource languages, speech/video/text/code modalities
Real-life Applications, e.g. text/speech/document translation, question-answering, coding, dialogue systems

Highlights

Our multi-agent approach to machine translation of long context (TransAgents) is covered in Andrew NG's newsletter
Our team presented 3 papers/posters + 1 tutorial at ACL 2024! You can obtain the slides and resources of the tutorial on Presentation Matters: How to Communicate Science in the NLP Venues and in the Wild?

Recent Grants and Projects

Trustworthy Generative AI: Towards Safe and Aligned Foundation Models, 840K, 2024-2026
Aligning Large Language Models with Human Intention, EBay Research, 130K, 2024-2027
Exploiting Context in Multilingual Understanding and Generation, ARC Future Fellowship, 850K, 2020-2025
Hierarchical Abstractions and Reasoning for Neuro-Symbolic Systems, 5.5M, 2023-2027
Large-scale multimodal knowledge management: From organization and user modeling to fast contextual presentation, 4M, 2022-2025
Dialogue Assistance for Negotiations in Cross-cultural Settings: A Neuro-Symbolic Computational Approach, 3.4M, 2022-2024
Learning to Learn and Adapt with Less Labels, 3.2M, 2019-2022
Continual Active Learning, EBAy Research, 200K, 2022-2023
Document-Wide Neural Machine Translation, EBay Research, 130K, 2020-2021
Effective Multi-Task Learning for Neural Machine Translation, Amazon Faculty Research Award, 113K, 2019-2020
Effective Learning of Document Neural Machine Translation, Google Faculty Research Award, 100K, 2019-2020
Document Machine Translation as Deep Structured Prediction, Google Faculty Research Award, 120K, 2018-2019
Explaining the outcomes of complex models, 430K, ARC DP, 2018-2022
Learning Deep Semantics for Automatic Translation between Human Languages, ARC DP, 450K, 2015-2018

Recent Professional Services

Reviewer, International Conference on Learning Representations (ICLR), 2025
Senior Area Chair, Empirical Methods in Natural Language Processing (EMNLP), 2024
Reviewer, Neural Information Processing Systems (NeurIPS), 2024
Reviewer, Association for Computational Linguistics (ACL), 2024


Gholamreza (Reza) Haffari Professor, ARC Future Fellow Monash University, Australia E:gholamreza.haffariATmonash.edu T:+61 3 9905-8106 Home Research Publications Research Team Projects & Areas Talks & Tutorials Data & Codes Teaching Contact Disclaimer: This page, its content and style, are the responsibility of the author and do not necessarily represent the views, policies or opinions of Monash University.	Positions Director of Engagement, Department of Data Sciece and AI, Monash University, 2024-now Head of the Language and Speech Processing, OpenStream AI, 2022-now. ARC Future Fellow and Professor, Department of Data Sciece and AI, Monash University, 2022-now Director of the Vision and Language Group, Department of Data Sciece and AI, Monash University, 2020-2024. Prospective Students For all prospective students (undergrad/msc/phd), please read this before you contact me. Due to the large volume of emails, I may not be able to answer all emails; apologies! To international undergrad students: Please email me only if you have your own funding for internship as I do not have any funding to support interns. Research Areas Neuro-Symbolic AI Agents, e.g. integrating LLMs and symbolic methods for multi-modal agents that operate/learn by perception, planning, acting, and feeback Machine Learning, e.g. deep generative models, Bayesian learning and inference, deep causal models, explainable AI, deep reinforcement learning Knowledge Graphs and LLMs, e.g. effective approaches for hybrid LLM-KG reasoning, certifying LLM knwoledge/reasoning Efficient and effective LLMs, e.g. better architectures, better decoding/inference, better training algorithms (long-context LLMs, KV-caching) Continual Learning, e.g. LLM knwoledge editing, retreival augmented generation (RAG) Reliable and Safe LLMs, e.g. alignment algorithms (RLHF), social-cultural norms, jailbreaking and defence mechanisms Multilingual and Multimodal Foundation Models, e.g. covering low-resource languages, speech/video/text/code modalities Real-life Applications, e.g. text/speech/document translation, question-answering, coding, dialogue systems Highlights Our multi-agent approach to machine translation of long context (TransAgents) is covered in Andrew NG's newsletter Our team presented 3 papers/posters + 1 tutorial at ACL 2024! You can obtain the slides and resources of the tutorial on Presentation Matters: How to Communicate Science in the NLP Venues and in the Wild? Recent Grants and Projects Trustworthy Generative AI: Towards Safe and Aligned Foundation Models, 840K, 2024-2026 Aligning Large Language Models with Human Intention, EBay Research, 130K, 2024-2027 Exploiting Context in Multilingual Understanding and Generation, ARC Future Fellowship, 850K, 2020-2025 Hierarchical Abstractions and Reasoning for Neuro-Symbolic Systems, 5.5M, 2023-2027 Large-scale multimodal knowledge management: From organization and user modeling to fast contextual presentation, 4M, 2022-2025 Dialogue Assistance for Negotiations in Cross-cultural Settings: A Neuro-Symbolic Computational Approach, 3.4M, 2022-2024 Learning to Learn and Adapt with Less Labels, 3.2M, 2019-2022 Continual Active Learning, EBAy Research, 200K, 2022-2023 Document-Wide Neural Machine Translation, EBay Research, 130K, 2020-2021 Effective Multi-Task Learning for Neural Machine Translation, Amazon Faculty Research Award, 113K, 2019-2020 Effective Learning of Document Neural Machine Translation, Google Faculty Research Award, 100K, 2019-2020 Document Machine Translation as Deep Structured Prediction, Google Faculty Research Award, 120K, 2018-2019 Explaining the outcomes of complex models, 430K, ARC DP, 2018-2022 Learning Deep Semantics for Automatic Translation between Human Languages, ARC DP, 450K, 2015-2018 Recent Professional Services Reviewer, International Conference on Learning Representations (ICLR), 2025 Senior Area Chair, Empirical Methods in Natural Language Processing (EMNLP), 2024 Reviewer, Neural Information Processing Systems (NeurIPS), 2024 Reviewer, Association for Computational Linguistics (ACL), 2024