Can Language Models Play the Wikipedia Game?
GPT-4o, o3-mini, QwQ-32B, and DeepSeek-R1-70B matched humans at Wikipedia-link navigation without clearly beating them.
1 stories, newest first · one card per story
GPT-4o, o3-mini, QwQ-32B, and DeepSeek-R1-70B matched humans at Wikipedia-link navigation without clearly beating them.
Today 2.1K readers · 2.6K views · 388 reads · all time 108K views