Woofun AI reports that the DMind Benchmark, developed by the Minara team, has been officially accepted into the Datasets & Benchmarks Track of KDD 2026. This inclusion establishes it as the first cross-disciplinary large-scale model evaluation benchmark at the intersection of digital assets and AI to undergo peer review for a top international conference, distinct from workshops or demos. KDD remains a premier venue in data mining and machine learning, with this track maintaining an acceptance rate of approximately 29%.

The benchmark assesses multi-step reasoning in real-world digital asset scenarios through 3,154 objective questions and 389 expert-reviewed open-ended tasks across nine major subfields, including DeFi, Tokenomics, and smart contract security. It has evaluated 31 leading models such as GPT, Claude, and Gemini. The dataset previously reached number one on Hugging Face's Trending list with over 13,000 cumulative downloads. A July 2026 reevaluation indicated that recent flagship models have not closed significant capability gaps, highlighting the benchmark's role in enabling AI to truly understand the world of digital assets.