Associate Professor · EEE · NTU Singapore

নমস্কার I am Soujanya Poria সৌজন্য পড়িয়া

I lead DeCLaRe Lab at NTU Singapore.

Research themes

Our lab works across six research themes: Safety, Trustworthiness, Multimodality, AI for Science, Efficiency, and Embodied AI. We pursue this work at the DeCLaRe Lab, NTU Singapore.

Work with our lab

Prospective students and collaborators can explore our research slides and recent publications to find questions that connect with their interests.

Research slides

Recent updates

Award

Highly Cited Researcher

Recognized by Web of Science.

Media

VentureBeat features δ-mem

An interview on working memory for long-running AI agents.

Talk

Healthcare AI Symposium

Invited master class in Singapore.

NTU

Joined EEE, NTU

Moved to the School of Electrical and Electronic Engineering.

More academic activities

DeCLaRe Lab

Much of this work grows out of the DeCLaRe Lab. Founded at SUTD in 2019, the lab has been based at NTU since 2025.

Selected contributions

Selected recent and foundational work.

Complete publication record
Current directions
2026Memory · arXiv

Delta-Mem

Online memory for language-model agents, enabling experience to shape future behavior without retraining the base model.

2026Efficiency · ICML

Data-Agent

Reframes data selection as an adaptive process in which models learn which training examples they need.

2026Trustworthiness · arXiv

Epistemic Context Learning

Gives language-model agents explicit epistemic context for deciding what to trust in multi-agent communication.

2026Generative audio · ICLR

TangoFlux

Fast, high-fidelity text-to-audio generation through flow matching and diffusion-transformer modeling.

2026Safety · ICLR

OffTopicEval

Tests whether language models can recognize when a request falls outside the active conversational context.

2025Trustworthiness · ICLR

Trustworthy RAG

Studies when retrieval strengthens language models and when conflicting evidence makes their answers less reliable.

2025AI for Science · ICLR

MOOSE-Chem

Evaluates whether language models can rediscover previously unseen chemistry hypotheses from scientific evidence.

2025Efficiency · arXiv

Error-Free Linear Attention

Derives exact linear-attention dynamics that improve efficiency without introducing approximation error.

2025Embodied AI · arXiv

NORA-1.5

Trains a generalist vision-language-action model with preference rewards grounded in predicted worlds and executed actions.

Foundational contributions
2024Trustworthiness · ICLR

Chain-of-Knowledge

Grounds language models by adapting knowledge dynamically across heterogeneous sources during reasoning.

2023Safety · arXiv

Chain of Utterances

Introduces multi-turn adversarial conversations for red-teaming the safety alignment of language models.

2021Trustworthiness · Cognitive Computation

Gender Bias in BERT

Provides a systematic account of gender bias encoded by contextual language representations.

2020Multimodality · ACM MM

MISA

Separates modality-specific and modality-invariant representations for robust multimodal learning.

2020Affective AI · EMNLP Findings

COSMIC

Integrates commonsense knowledge into contextual emotion recognition for natural conversations.

2019Affective AI · EMNLP

DialogueGCN

Represents speakers and conversational dependencies as a graph for emotion recognition in dialogue.

2019Affective AI · AAAI

DialogueRNN

Models conversational context and speaker state for emotion recognition in multi-party dialogue.

2019Affective AI · ACL

MELD

Established a multimodal, multi-party benchmark for emotion recognition in natural conversations.

2017Multimodality · EMNLP

Tensor Fusion Network

Introduced explicit tensor-based interactions among language, vision, and acoustic representations.