Detecting Large Language Models via Human Memory Limits
Learn how probing human working memory constraints helps distinguish large language models from humans in online behavioral research.
MSA-Thinker: Advanced Reinforcement Learning for Multimodal Sentiment
Discover MSA-Thinker, a novel framework using hint-guided reinforcement learning to improve multimodal sentiment analysis with better accuracy and interpre...
Reactivating Hidden Safety in Post-Trained Large Language Models
Discover how SafeReAct restores suppressed safety mechanisms in post-trained LLMs, enhancing security without losing reasoning power.
Gender Bias in ChatGPT Hiring: Study & Solutions
Explore how ChatGPT exhibits gender bias in hiring and how prompt engineering can help reduce these biases for fairer AI-driven recruitment.
Can Large Language Models Accurately Perceive Time?
Explore how large language models struggle to estimate task durations accurately, impacting scheduling and planning in AI applications.