Tej Deep Pala

I am currently a 1st year PhD student at Nanyang Technological University (NTU), advised by Prof. Soujanya Poria. My research focuses on Large Language Model Reasoning, and Reinforcement Learning.

Previously, I earned my Bachelor of Engineering in Computer Science and Design from Singapore University of Technology and Design (SUTD) in May 2024 with Highest Distinction.

If you are interested in collaborating or chatting, feel free to reach out to me at ptejdeep@gmail.com.

News

[Jan 2026] Our paper KAIROS has been accepted to ICLR 2026!
[Aug 2025] Began my PhD journey at Nanyang Technological University (NTU)!
[Aug 2025] Two papers, FERRET and PathFinder-PRM, have been accepted to EMNLP Findings 2025.
[May 2025] Emma-X has been accepted to ACL Main 2025.
[May 2024] Graduated from SUTD with a B.Eng. in Computer Science and Design with Highest Distinction.

Publications

Measuring and Mitigating Rapport Bias of Large Language Models under Multi-Agent Social Interactions

Maojia Song*, Tej Deep Pala*, Ruiwen Zhou, Weisheng Jin, Amir Zadeh, Chuan Li, Dorien Herremans, Soujanya Poria

ICLR Poster, 2026

[arXiv] [OpenReview] [ICLR]

Error Typing for Smarter Rewards: Improving Process Reward Models with Error-Aware Hierarchical Supervision

Tej Deep Pala, Panshul Sharma, Amir Zadeh, Chuan Li, Soujanya Poria

EMNLP Findings, 2025

[arXiv] [ACL Anthology]

FERRET: Faster and Effective Automated Red Teaming with Reward-Based Scoring Technique

Tej Deep Pala*, Vernon Y.H. Toh*, Rishabh Bhardwaj, Soujanya Poria

EMNLP Findings, 2025

[arXiv] [ACL Anthology]

Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning

Qi Sun, Pengfei Hong, Tej Deep Pala, Vernon Y.H. Toh, U Tan, Deepanway Ghosal, Soujanya Poria

ACL Main, 2025

[arXiv] [ACL Anthology]

Preprints

Training Vision-Language Process Reward Models for Test-Time Scaling in Multimodal Reasoning: Key Insights and Lessons Learned

Brandon Ong, Tej Deep Pala, Vernon Toh, William Chandra Tjhi, Soujanya Poria

[arXiv]

Lessons from Training Grounded LLMs with Verifiable Rewards

Shang Hong Sim*, Tej Deep Pala*, Vernon Toh*, Hai Leong Chieu, Amir Zadeh, Chuan Li, Navonil Majumder, Soujanya Poria

[arXiv]

PromptDistill: Query-based Selective Token Retention in Intermediate Layers for Efficient Large Language Model Inference

Weisheng Jin, Maojia Song, Tej Deep Pala, Yew Ken Chia, Amir Zadeh, Chuan Li, Soujanya Poria

[arXiv]

DELLA-Merging: Reducing Interference in Model Merging through Magnitude-Based Sampling

Tej Deep Pala, Rishabh Bhardwaj, Soujanya Poria

[arXiv]