Research map / Label-efficient and robust learning / Self-supervised and contrastive
From Faults to Features: Pretraining to Learn Robust Representations against Sensor Failures
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes pretraining to learn representations robust to sensor failures, targeting safety-critical perception such as autonomous driving.
Problem
Sensor faults corrupt input signals in safety-critical applications (autonomous vehicles, ADAS), causing severe model failures at inference time.
Approach
Uses pretraining to learn representations that remain robust to sensor failures, rather than relying only on fault-specific fine-tuning or augmentation; the abstract's details beyond this framing are not stated.
Closest papers by meaning
- Harnessing Event Sensory Data for Error Pattern Prediction in Vehicles: A Language Model Approach (AAAI 2025 · similarity 0.86)
- CaDeT: a Causal Disentanglement Approach for Robust Trajectory Prediction in Autonomous Driving (CVPR 2024 · similarity 0.84)
- Monitoring of Perception Systems: Deterministic, Probabilistic, and Learning-Based Fault Detection and Identification (Abstract Reprint) (AAAI 2024 · similarity 0.83)
- SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models (ICML 2025 · similarity 0.82)
- Understanding and Enhancing Mask-Based Pretraining towards Universal Representations (NeurIPS 2025 · similarity 0.82)
- Progressive Risk Estimation for Accident Anticipation (NeurIPS 2026 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.