Most people on Earth speak more than one language and often switch languages in the same chat, but AI tools aren’t tested well on this real behavior.
Large reasoning models can often find the right math answer in their “head” before finishing their written steps, but this works best in languages with lots of training data like English and Chinese.
ProGuard is a safety guard for text and images that doesn’t just spot known problems—it can also recognize and name new, never-seen-before risks.
EditThinker is a helper brain for any image editor that thinks, checks, and rewrites the instruction in multiple rounds until the picture looks right.
The paper shows how a vision-language model (VLM) can train itself to be a fair judge of answers about images without using any human preference labels.