← Search

Yiwei Liao

2 accepted papers

2026

TruthRL: Incentivizing Truthful LLMs via Reinforcement Learning

ICML 2026poster

While large language models (LLMs) have demonstrated strong performance on factoid question answering, they are still prone to hallucination and untruthful responses, particularly when tasks demand information outside their parametric knowledge. Indeed, truthfulness requires more than accuracy---mod…

Cited by 0SourceScholar
2025

WearVQA: A Visual Question Answering Benchmark for Wearables in Egocentric Authentic Real-world scenarios

NeurIPS 2025poster

We introduce WearVQA, the first benchmark specifically designed to evaluate the visual question answering (VQA) capabilities of multi-modal AI assistant on wearable devices like smart glasses. Unlike prior benchmarks that focus on high-quality, third-person imagery, WearVQA reflects the unique chal-…

Cited by 0SourceScholar