3 papers
cs.AI2026
Causal Attribution for Agentic Decisions: Estimators, Coupling, and a Traceability Specification
Ajay Pravin Mahale
A provider of a high-risk AI system must keep records that make a decision traceable, and for agentic systems it has not been established what those records must contain for post-h…
cs.AI2026
Explanation Multiplicity: Circuit-Level Interpretability Evidence Does Not Survive Defensible Analytic Variation
Ajay Pravin Mahale
The EU AI Act requires providers of high-risk systems to file technical documentation describing how the system reaches its decisions. Mechanistic interpretability is the obvious s…
cs.CL2026
Causally Grounded Mechanistic Interpretability for LLMs with Faithful Natural-Language Explanations
Ajay Pravin Mahale
Mechanistic interpretability identifies internal circuits responsible for model behaviors, yet translating these findings into human-understandable explanations remains an open pro…