> As part of our IMO work, we also experimented with a natural language reasoning system, built upon Gemini and our latest research to enable advanced problem-solving skills. This system doesn’t require the problems to be translated into a formal language and could be combined with other AI systems. We also tested this approach on this year’s IMO problems and the results showed great promise.
Wonder what "great promise" entails. Because it's hard to imagine Gemini and other transformer-based models solving these problems with reasonable accuracy, as there is no elimination of hallucination. At least in the generally available products.
They explicitly stated that to achieve the current results, they had to manually translate the problem statements into formal mathematical statements:
> First, the problems were manually translated into formal mathematical language for our systems to understand.
How I understand what they're saying is that they used gemini to translate the problem statement into formal mathematical language and let DeepMath do it's magic after that initial step.
Wonder what "great promise" entails. Because it's hard to imagine Gemini and other transformer-based models solving these problems with reasonable accuracy, as there is no elimination of hallucination. At least in the generally available products.