{"id":18387,"date":"2024-08-21T09:29:19","date_gmt":"2024-08-21T01:29:19","guid":{"rendered":"https:\/\/www.1ai.net\/?p=18387"},"modified":"2024-08-21T09:28:44","modified_gmt":"2024-08-21T01:28:44","slug":"testtt","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/18387.html","title":{"rendered":"Tongyi Qianwen Mathematical Model Qwen2 Math Demo released, 72B version beats GPT-4"},"content":{"rendered":"

Ali Baba's.\"Thousand Questions on Tongyi<\/a>\"The team has another big story! They just released Qwen2Math DemoMathematical model<\/a>It's just a little monster.GPT-4<\/a>All were trampled under its feet.<\/p>\n

This model can not only handle math problems entered in text, but also understand formulas in pictures and screenshots. Imagine that you take a picture of a math equation and it can give you the answer. It is simply a magic tool for solving math problems in math class! (Of course, we do not encourage cheating)<\/p>\n

\"fd2d57f8j00sijo2k001cd0015p00kym\"<\/p>\n

Qwen2-Math has launched three versions: 72B, 7B and 1.5B. The 72B version is simply a math genius. It scored 7 points more than GPT-4 on the MATH dataset, an improvement of 9.6%. This is like you scored 145 points in the college entrance examination math, while the top student next to you only scored 132 points.<\/p>\n

Worse still, version 7B uses less than one tenth of the parameter, exceeding the open source mathematical model NuminaMath of 72B. You know, NuminaMath is the model that won the award at the first AIMO in the world, and the prize was awarded by the top master of mathematics himself\u3002<\/p>\n

Ali's senior algorithm expert, Lin Joon, announced with excitement that they had turned the Qwen2 model into a mathematical master. How do they do that? They use a specially designed \"mathematical condensation\" -- a very well-designed mathematical language library. This \"embracing fluid\" contains a large number of high-quality mathematical web-based texts, books, codes, examination topics and even Qwen2 models themselves\u3002<\/p>\n

The result? In classic math test sets such as GSM8K and MATH, Qwen2-Math-72B left 405B's Llama-3.1 behind. These test sets are no joke, they contain algebra, geometry, probability, number theory and other math problems.<\/p>\n

Not only that, Qwen2-Math also challenged the Chinese dataset CMATH and college entrance examination questions. On the Chinese dataset, even the 1.5B version can beat the 70B Llama3.1. Moreover, no matter which version, the results are significantly improved compared with the Qwen2 basic model of the same scale.<\/p>\n

Looks like the Quaker asked a real math genius this time. Can we ask him a math question later? But remember, it's just a tool, but don't let his intelligence confuse your eyes!<\/p>\n

Online experience address: https:\/\/huggingface.co\/spaces\/Qwen\/Qwen2-Math-Demo<\/p>","protected":false},"excerpt":{"rendered":"

Ali Baba's \"Twilight Ask\" team made another big story. They just released Qwen2Math Demo, a mathematical model that is a little monster, and even GPT-4 is under its feet. This model not only addresses the mathematical aspects of text input, but also understands the formulas in the pictures and screenshots. Imagine, if you take a calculus picture, it'll answer you. It's a mathematician. Of these, 72B is a mathematical genius that scored seven points above GPT-4 in the MATH data set, an improvement of 9.6%. It's like you got 145 in G<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[510,4087,331],"collection":[],"class_list":["post-18387","post","type-post","status-publish","format-standard","hentry","category-news","tag-gpt-4","tag-4087","tag-331"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/18387","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=18387"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/18387\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=18387"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=18387"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=18387"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=18387"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}