Fine-tuning multimodal language models on a new SOTIF-focused driving dataset improves question answering and captioning, but the open-ended gains are measured by an LLM judge with no independent human scoring.
Pesotif: A challenging visual dataset for perception sotif problems in long-tail traffic scenarios,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DriveSOTIF: Advancing Perception SOTIF Through Multimodal Large Language Models
Fine-tuning multimodal language models on a new SOTIF-focused driving dataset improves question answering and captioning, but the open-ended gains are measured by an LLM judge with no independent human scoring.