We welcome you to CDS-KIAC talk on 06th July 2026 (Monday). The details are as below:
Speaker : Prof. Anoop Cherian, Senior Principal Research Scientist at (MERL)
Title : Toward Societies of Agents: Reasoning, Debate, and Collaborative Intelligence
Date and Time : July 06, 2026: 04:00 PM
Venue : # 102, CDS Seminar Hall
ABSTRACT:
The next frontier of artificial intelligence may not be a single monolithic model, but a society of agents that can reason, disagree, coordinate, and learn through interaction. While large language and vision-language models have shown remarkable capabilities, they remain fragile as isolated decision makers: they hallucinate, miscoordinate with partners, and often struggle to adapt to collaborative tasks. This talk asks a central question: how can agent interaction become a computational principle for building more reliable, reflective, and cooperative AI systems? I will present two complementary lines of work toward this goal. In WISE, we study reasoning as structured debate among heterogeneous foundation models. Rather than treating disagreement as noise, WISE organizes models into specialized roles—solvers, critics, and aggregators—and uses multi-round interaction with probabilistic solution aggregation to turn agent diversity and uncertainty towards robust multimodal inference. In LLawCo, I will extend this perspective from debate to embodied cooperation. In decentralized, partially observable environments, agents must not only reason correctly, but also coordinate their actions and communication with partners. LLawCo learns high-level behavioral laws of cooperation from failed interactions, transforming recurring miscoordination into explicit priors for future planning, communication, and action. Together, these works advance a broader technical perspective: moving from foundation models as isolated predictors to agentic societies that reason through mutual interaction. Across multimodal reasoning and embodied collaboration benchmarks, we show that this structured interaction substantially improves robustness and achieves state-of-the-art performance, pointing toward AI systems that are not merely larger, but more collaborative, adaptive, and reliable.
BIOGRAPHY:
Anoop Cherian is a Senior Principal Research Scientist at Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, and an Adjunct Associate Professor at the Australian National University (ANU), Canberra, Australia. He was previously a Research Fellow in the Australian Centre for Robotic Vision (ACRV) at ANU and a Postdoctoral Fellow in the LEAR project team at Inria, Grenoble, France. He received his M.S. and Ph.D. degrees from the University of Minnesota, Twin Cities, in 2010 and 2013, respectively. He has authored more than 100 research papers and holds over 25 patents. He has organized several workshops, including serving as a lead organizer of the Multimodal Algorithmic Reasoning workshops at CVPR and NeurIPS over the past three years. He has also served as an Area Chair for top-tier computer vision and machine learning conferences since 2023. His current research focus includes multimodal foundation models, embodied multi-agent behavior modeling and orchestration, physics-grounded imagination, and diffusion-based generative approaches for building AI systems that can reason about, simulate, and act in the physical world.
Host Faculty: Prof. Venkatesh Babu, CDS
ALL ARE WELCOME
