Accepted Papers

• Accuracy Is Not Capability: A Causal Audit Standard for RF Classification

Atul Rawal and Adrienne Raglin

• Metacognition and Cognitive Transparency for Trustworthy Human-Robot Teams

Sanjay Oruganti, Sergei Nirenburg, Marjorie McShane and Jesse English

• Degradation-Resistant Post-Training: Mitigating Generative Degradation in Autoregressive Transformers via DI-Guided Optimization

Ritvik Garimella, Riju Marwah, Atishay Jain, Khusham Bansal and Amit Sheth

• On Robust and Efficient Chatbots and How to Automatically Compose Them from AI Components Using Automated Planning

Kausik Lakkaraju, John Aydin, Sai Teja Paladi, Sunandita Patra, Biplav Srivastava and Parisa Zehtabi

• Structural Metric Evaluation of LLM Generated Ontologies

Ryn Gray, Jaimie Murdock, Douglas C. Crowder and Jarod Kaltenbaugh

• Hybrid Artificial Intelligence for Critical Systems: Combining Data-Driven and Knowledge-based Approaches for Validity, Reliability, Robustness, and Explainability

Juliette Mattioli, Christophe Guettier and Nicolas Museux

• Uncovering the Security Debt of Machine Unlearning in Agentic Deployment

Md Khairul Azman and Shibbir Ahmed

• Context Matters: Evaluating LLM-Generated Knowledge Graph Schemas

Ritvik Garimella, Riju Marwah, Atishay Jain, Khusham Bansal and Amit Sheth

• From Accuracy to Assurance: Toward Credible Power Grid Anomaly Detection

Elisa Zhang, Lily Ridley and Michael Darling

• Soft-Label Governance for Distributional Safety in Multi-Agent Systems

Aizierjiang Aiersilan and Raeli Savitt

• Use Subject Matter Expertise “On-the-Loop”, not “In-the-Loop”: a Proposed Assurance Method and an Unresolved Question

Nathan Bos, Tonia Korves, Colin Diggs, Matthew Peterson, Laurie Damianos, Rebecca Chance, Aaron Giera, Purushotham Karnam, Saleem Mohammed and Saiju Pyarajan

• An Exploration of Abstention and Calibration in Assurance of Multi-class Classifiers

David Nahmias, Ariel Kapusta, Pam Bhattacharya and Patrick Minot

• The Decisiveness Trap: Inverse Correlation Between Performance and Socioeconomic Bias in Frontier LLMs

Vibha Santhanakrishnan

• Geometry-Based Refinement of Operational Design Domains for Physics-Constrained AI-based Systems arising from Riemannian Flow Matching

Binh Tran, Martin Gonzalez, Frédéric Barbaresco and Florence D'Alché-Buc

• xLightGCN: Explainable Graph Neural Network Based Recommendation Systems with Most Influential Peers

Calder Johnson and Robin Cohen

• From Severity to Consequence: A Measurement Agenda for Securing AI Decisions in Seafood Safety

Lihong Yang, Ran Yang and Yiming Feng

• Accountable Delegation: A Layered Governance Framework for Trustworthy Multi-Agent AI in Risk-Averse Domains

Nikhil Mittal

• A Neurosymbolic Architecture for Establishing Trust in Agentic Advertising: Lessons from an Early Instantiation

Satyajeet Raje, Jose David Aguas Lopes, Piotr Kostrzeński and Tiago Ramos

• RAISE: A Novel Approach to Assure the Safety of Vision–Language-Action-Based Driving Systems

Gerhard Yu, Fuyuki Ishikawa, Oluwafemi Odu and Alvine B. Belle

• ATM-MEMORYTRAP: A Benchmark for Auditing Action Transferability Beyond Semantic Similarity in Agentic Air Traffic Management

Soufiane Momtaz

• CRUCIBLE: A Multi-Agent Framework for Automated Annotation Rubric Calibration via Iterative Inter-Judge Disagreement Resolution

Bidit Das

• The Defense Rests: A Courtroom-Inspired Multi-Agent LLM Framework for Automated Safety Case Generation

Gerhard Yu, Amanuel Aknaw, Amir Mohaghegh, Yong Song Cao and Alvine B. Belle

• Measuring Source Fidelity, Exculpatory Recall, and Generative Suspicion in AI-Drafted Police Reports

Aizierjiang Aiersilan and Yeershanati Wuernikebai

• Low-Rank Bayesian Structured Adaptation of Large Language Models

Deepak Kandel and Dimah Dera

• Bring Your Own Assurance: Extensible, Composable AI Test and Evaluation with CheckMAITE

Kenneth Foster, Nick Byrne, Brandon Geraci, Mike Mazara, Oren Fromberg, Kimberly Pevey and Dharhas Pothina 

• Mechanism Matters: Modality Informativeness and Fusion Routing Determine Missing-Modality Robustness in Multi-Modal Federated Learning

Atul Rawal and Caleb Cooper

• Uncovering Default Gender Biases in Publicly Accessible LLMs Through Counter-Balancing Cloze-Task (CBCT)

Faouzi Adjed, Lucas Schott, Lucas Mattioli, Jaime De Oliveira, Emna Amdouni and Sabrina Chaouche

• From Incredible Claims to Credible AI: Evidence-Backed Maturity Profiles for Risk-Informed AI Assurance

Jake Nichol and Michael Darling

• Assurance Properties of Shielding Techniques for Safe Reinforcement Learning: A Comparative Analysis

Srinivas Nedunuri

• A Social Choice Theory Approach towards Quantifying Causal Reasoning in Large Language Models

Elias Boshara, Michael Darling and Reed Milewicz

• Arguing Residual Risk Acceptance for E2E AI Automated Driving Systems using Black-box Risk-based Testing

Rasmus Adler, Ioannis Sorokos, Jan Reich, Raphael Fonte Boa Trindade, Sandro Nueesch and Thomas Muehlenstaedt

• Auditing the Auditor: Calibrated, Label-Free Measurement of Ungrounded Claims in LLM Incident Diagnosis

Yash Rajeshbhai Parikh

• Fairness Is Not Universal: Towards Contextualized and Autonomous Fairness in Federated Learning

Yi Zhou and Naman Goel

• Publish the Disagreement: Instrumented Human Verification in an Agentic Research Pipeline

Jiajun Ma

• Auditing Evidence Claims in Federal High-Impact AI Exclusions

Hema Raju Barri and Venkateswarlu Nagineni

• From Agreement to Action: A Reliability Contract for Multi-Judge LLM Assurance

Khrystyna Terletska

• Trust Without the Word: Object-Scoped Evaluation and the Case for an Operational Design Domain for Trust Metrics

Ali Shahidy and Usha Lakshmanan