Chenhang_AI@cuichenhangOct 3, 2023New research! We are excited to announce our latest investigation into object hallucinations in Language-Visual Linking Models (LVLMs) and our collaborative efforts with @AiYiyangZ, @HuaxiuYaoML, @jaeh0ng_yoon, @linjunz_stat, @zhun_deng, @chelseabfinn, @mohitban47.Huaxiu Yao@HuaxiuYaoMLOct 3, 2023Replying to @HuaxiuYaoML🔬Our solution stems from rigorous statistical analysis pinpointing the causes of hallucination in LVLMs. We've identified three pivotal factors: - Object co-occurrence - Object uncertainty - Position of the object in generated descriptions27714
Huaxiu Yao@HuaxiuYaoMLOct 3, 2023Replying to @HuaxiuYaoML🔬Our solution stems from rigorous statistical analysis pinpointing the causes of hallucination in LVLMs. We've identified three pivotal factors: - Object co-occurrence - Object uncertainty - Position of the object in generated descriptions
