Prof. Karthik Vaidhyanathan Presents Research on GenAI-Based Adaptive Systems at ACSOS 2026

Prof. Karthik Vaidhyanathan Presents Research on GenAI-Based Adaptive Systems at ACSOS 2026

Prof. Karthik Vaidhyanathan of IIIT Hyderabad’s Software Engineering Research Center (SERC) presented the paper “Mitigating Uncertainty Interactions in GenAI-based Adaptive Systems: Vision, Challenges and Preliminary Guidelines” at the IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS 2026), held in Cesena, Italy, on 9 September 2026. The paper was co-authored with researchers from international institutions.

As software-intensive systems increasingly integrate Generative AI (GenAI), large language models (LLMs), and agentic components, they face new and interacting sources of uncertainty, including unpredictable outputs, hallucinations, context-window limitations, and outdated memory. These uncertainties can affect system reliability, quality, and runtime behaviour.

The paper proposes a conceptual framework and preliminary guidelines for identifying and mitigating interactions among uncertainties throughout the software lifecycle, from requirements and architectural design to runtime monitoring and adaptation. It also illustrates the approach through three use cases: LLM serving, AI-driven software operations, and agentic cloud management.

The work contributes to research on dependable GenAI-based adaptive systems by highlighting the need to address interacting uncertainties as part of software design and runtime adaptation.