RASACC 2026

Workshop Program June 22, 2026

Program

The RASACC 2026 program features keynote talks, technical paper presentations, and a hands-on session on resilience, trust, and assurance for autonomous and agentic systems.

9:00–10:15

Keynote

Keynote talk by Jérémie Decouchant

Title

Using Trusted Execution Environments to accelerate and reinforce blockchain consensus.

Bio and Abstract

Jérémie Decouchant is an assistant professor at TU Delft, specializing in the design of robust distributed systems and algorithms. His research focuses on mechanisms that prevent, tolerate, detect, and remove faults in multi-party environments. He has been applying these dependable computing principles to blockchain, healthcare, and learning systems.

The presentation will discuss recent breakthroughs in streamlined BFT consensus algorithms. Specifically, it will cover how leveraging trusted components can reduce communication phases and replication degrees (EuroSys'22, IPDPS'24, TDSC'24), how this approach surprisingly yielded a novel hierarchical consensus algorithm (INFOCOM'24), and how to defend against rollback attacks when trusted components lose their state (NDSS'26).

10:15–10:30

Technical session

Session I: Foundations of Resilient Agentic Systems

(RASACC) Automated Resilience Assessment for Autonomous System-based System-on-Chip Architectures

Abigail Butka and Brian Butka

10:30–11:00

Break

Coffee break

11:00–12:30

Technical session

Session II: Secure and Adaptive Agentic Systems

(RASACC) DisTrust: Adaptive Trust-Based Verification for Resilient Distributed Computation

Fouad Afiouni, Saurabh Bagchi, and Aravind Machiry

(CRAI) From Agent Failure Paths to Quantified Residual Risk: A Compositional Framework for Resilient Agentic AI

Hassan Karim, Deepti Gupta, Sai Sitharaman, and Danda Rawat

(CRAI) Agentic AI for Continuous Intelligence in Future Smart Connected Systems

Likhitha Annapurna Kavuri, Deepti Gupta, and Eman Hammad

(CRAI) Multi-Agent-based Failure Recovery Planning for a Connected Vehicle Platform

Aoi Matsuda, Ippo Hiroi, Fumio Machida, and Naohiro Ohtsuka

(CRAI) Composable Trust in Agentic AI: Bridging Architectural Capability and System-Level Assurance

Omar Al-Refai, Ibrahim Shahbaz, and Eman Hammad

12:30–14:00

Break

Lunch break

14:00–15:15

Keynote

Keynote talk by Ehsan Aghaei

Title

TBD

Bio and Abstract

TBD

TBD

15:15–15:30

Technical session

Session III: Defensive Agentic Skills

(CRAI) NinjaShell: Skill-Aware Defense Against Malicious Agentic Skills

Haddy Alchaer, Joseph Ghanem, and Saurabh Bagchi

15:30–16:00

Break

Coffee break

16:00–17:00

Keynote

Keynote talk by P.R. Kumar

Title

TBD

Bio and Abstract

TBD

TBD

17:00–17:30

Hands-on

Hands-on session