teLLMe Why (Ain’t Nothing but a Jam): Exploratory Causal Analysis of Urban Driving Data

Qiwei Li, Jorge Ortiz

Keywords: Causal Inference, Urban Computing, Computer Vision, Causal Discovery, Explainable AI, Machine Learning, Urban Video Analytics

TL;DR: A system that turns dashcam-derived traffic events into interactive causal “what-if” analyses, using a learned DAG and LLMs to estimate effects and explain them in plain language.

Abstract:

Traffic agencies now have access to large volumes of video-derived data for study-
ing safety and congestion. Most of these data are observational and collected with-
out interventions, which makes causal questions such as “How would rain change
traffic density?” difficult to answer. We present teLLMe, a system for exploratory
causal analysis of urban driving datasets. The system starts from a structured
event table built from dashcam annotations and combines causal structure learn-
ing with the PC algorithm, bootstrap-based stability checks, and query-specific
effect estimation using linear regression and DoWhy. Natural-language questions
are mapped to structured causal queries through a schema-aware LLM, enabling
users to specify treatments, outcomes, and subpopulations. teLLMe returns a
“Causal Card” that summarizes effect estimates, adjustment sets, DAG support,
and assumptions, followed by a short natural-language explanation. Case studies
on BDD-derived traffic events show that the system can surface plausible relation-
ships involving weather, peak hours, and traffic density, while making uncertainty
and modeling choices explicit. The system is designed as a tool for hypothesis
generation and expert reasoning rather than a source of definitive causal claims.

Submission Number: 37