aiqbench analysis
NOVA demonstrates that autonomous symbolic regression can extract interpretable, high-fidelity models of car-following and lane-changing directly from raw trajectory data without relying on heavy behavioral priors. By identifying a dominant nonlinear term that generalizes zero-shot across different freeway sites, the framework bridges the gap between complex trajectory prediction and established psychophysical theories of collision avoidance. This approach offers a robust alternative to black-box baselines, particularly for applications requiring explainability and transferability across diverse driving environments.
NOVA: Symbolic Regression Discovery of Interpretable Car-Following and Lane-Change Models with Driver Heterogeneity
arXiv:2606.10583v1 Announce Type: new Abstract: We present NOVA, an autonomous symbolic regression framework that identifies interpretable car-following and lane-change structures from raw trajectory data with minimal behavioral priors. Applied to 4,765,788 active driving observations from the NGSIM I-80 and US-101 datasets, NOVA's deterministic Rust-powered search engine evaluates over 10,000 candidate algebraic structures and identifies a compact two-term acceleration model under a forward-shifted rolling-mean prediction target. Evaluated under two complementary preprocessing pipelines, NOVA achieves $RMSE = 1.376 m/s^2$ ($R^2 = 15.57\%$) on the intent-forecasting benchmark, outperforming the best recalibrated symbolic-regression baseline (SR-LLM, PNAS~2025) by 0.135 m/s$^2$ in RMSE under an identical evaluation protocol. Across eight independent experiments, a single dominant nonlinear term emerges as a robust backbone of human car-following; a residual-guided extension further links the selected structure to an established psychophysical theory of collision avoidance. The discovered feature operators transfer zero-shot between freeway sites with under 3 pp $R^2$ loss. Extended to lane-change modelling within a multinomial logit framework, NOVA achieves 67.4\% balanced accuracy under strict vehicle-ID holdout on 502 unseen drivers, surpassing existing lane-changing baselines by +29.8 percentage points on a three-class problem.
Abstract:We present NOVA, an autonomous symbolic regression framework that identifies interpretable car-following and lane-change structures from raw trajectory data with minimal behavioral priors. Applied to 4,765,788 active driving observations from the NGSIM I-80 and US-101 datasets, NOVA's deterministic Rust-powered search engine evaluates over 10,000 candidate algebraic structures and identifies a compact two-term acceleration model under a forward-shifted rolling-mean prediction target. Evaluated under two complementary preprocessing pipelines, NOVA achieves $RMSE = 1.376 m/s^2$ ($R^2 = 15.57\%$) on the intent-forecasting benchmark, outperforming the best recalibrated symbolic-regression baseline (SR-LLM, PNAS~2025) by 0.135 m/s$^2$ in RMSE under an identical evaluation protocol. Across eight independent experiments, a single dominant nonlinear term emerges as a robust backbone of human car-following; a residual-guided extension further links the selected structure to an established psychophysical theory of collision avoidance. The discovered feature operators transfer zero-shot between freeway sites with under 3 pp $R^2$ loss. Extended to lane-change modelling within a multinomial logit framework, NOVA achieves 67.4\% balanced accuracy under strict vehicle-ID holdout on 502 unseen drivers, surpassing existing lane-changing baselines by +29.8 percentage points on a three-class problem.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
arXiv:2606.10583 [cs.LG]
(or arXiv:2606.10583v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.10583 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 280 words
Related coverage
From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification
arXiv · Jun 16, 2026
From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification
arXiv · Oct 1, 2026
Clinically Interpretable Biological Age Estimation Using CatBoost and SHAP with Continuous and Categorical Features from NHANES
arXiv · Oct 16, 2026
Splitting the Difference: Interpretable Causal Forests for Treatment Effect Heterogeneity and Bias
arXiv · Sep 16, 2026