LiDAR-Hallu Benchmark

What is LiDAR-Hallu?

Evaluates LiDAR language model responses on 10,000 geometry-referenced questions across 150 nuScenes scenes, with strict parsing and diagnostic controls.

Released

2026-09-21

Evaluates

General AI, Geometric reasoning, Vision & 3D, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.24452

LiDAR-Hallu paper and code

• Paper https://arxiv.org/abs/2609.24452
• Code https://github.com/Awesome4D/4DMLLM_Hallucination_Bench

Which sources cite LiDAR-Hallu?

• LiDAR-Hallu

Related General AI, Geometric reasoning, Vision & 3D, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A Real-World CAD Segmentation Benchmark
• A3S-Bench
• AARRI-Bench