Search papers, labs, and topics across Lattice.
4
3
8
9
CNeVA reveals that smooth eligibility gates can significantly enhance agent controllability and realism in traffic simulations, outperforming traditional methods.
Curvature-aware model merging, previously intractable, can now be efficiently approximated in a low-rank subspace, provably outperforming flat-geometry methods and unifying prior spectral techniques.
By recasting the Hamilton-Jacobi-Bellman equation as a tractable Monte Carlo estimation, this work stabilizes physics-informed RL and unlocks its potential for high-dimensional control tasks.
Finally, a real-world testing platform that can rigorously evaluate full-stack VLM-integrated autonomous driving systems, offering configurable scenarios and closed-loop control.
