teLLMe Why (Ain't Nothing but a Jam): Exploratory Causal Analysis of Urban Driving Data
2026-07-16 • Artificial Intelligence
Artificial IntelligenceHuman-Computer Interaction
AI summaryⓘ
The authors created a system called teLLMe to help explore cause-and-effect questions using large amounts of traffic video data. Since most data is just observed without controlled experiments, it's hard to know if one thing causes another, like how rain affects traffic. Their system uses advanced methods to guess potential causes and effects, turning user questions into structured queries with the help of a language model. It then provides a summary explaining these possible relationships and the assumptions involved, mainly to help experts form new ideas rather than give final answers.
causal analysisPC algorithmbootstraplinear regressionDoWhycausal inferencedirected acyclic graph (DAG)schema-aware language modelobservational datatraffic video data
Authors
Qiwei Li, Jorge Ortiz
Abstract
Traffic agencies now have access to large volumes of video-derived data for studying safety and congestion. Most of these data are observational and collected without 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 learning 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 relationships 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.