TraIL-Det introduces transformation-invariant local features that capture neighborhood geometry while maintaining robustness to point density variations
Explore technology by industry and role
Newest additions first. Explore past and current notes by industry and role.
All notes
Newest additions first. Explore past and current notes by industry and role.
A tracking model generates pseudo-ground-truths from LiDAR frame sequences to enhance few-shot learning.
The DRIVE framework proposes a solution to improve interpretability consistency and stability in autonomous driving models.
Introduces a bit-plane-selective, sensitivity-guided scheme to protect AI weights without extra storage overhead
Introduces outlier-weighted layerwise pruning to reduce LLM computational load efficiently.
Scientific hypotheses are replaced with sets of mutually exclusive and exhaustive competing explanations to clarify interpretation
Supports differentiation through both inputs and curve coefficients, enabling differentiable parametric curve learning.
A novel scheduling strategy for efficiently deploying perception workloads on multi-chiplet AI accelerators
A characterized diffusion module combined with a spatial-temporal interaction network improves prediction accuracy in dynamic and heterogeneous traffic.
Interaction events are extracted and structured from naturalistic driving records using formal methods
PKRD-CoT enables MLLMs to mimic human thought processes step-by-step by aligning with perception, knowledge, reasoning, and decision-making in autonomous tasks.
Learning experience is analyzed by AI to generate adaptive, self-adjusting scaffolds during training.
Integrates chain-of-thought reasoning in MLLMs to enable end-to-end autonomous driving implementation
Curriculum learning progressively increases environmental complexity to improve safety and generalization in autonomous driving
A novel trajectory planner is proposed to enable safe and efficient headland turning in cluttered orchard environments.