Despite Flow Matching and diffusion models having emerged as powerful generative paradigms for continuous variables such as images and videos, their application to high-dimensional discrete data, such as language, is still limited. In this work, we present Discrete Flow Matching, a novel discrete flow paradigm designed specifically for generating discrete data. Discrete Flow Matching offers several key contributions: (i) it works with a general family of probability paths interpolating between source and target distributions; (ii) it allows for a generic formula for sampling from these probability paths using learned posteriors such as the probability denoiser (x-prediction) and noise-prediction (ϵ-prediction); (iii) practically, focusing on specific probability paths defined with different schedulers improves generative perplexity compared to previous discrete diffusion and flow models; and (iv) by scaling Discrete Flow Matching models up to 1.7B parameters, we reach 6.7% Pass@1 and 13.4% Pass@10 on HumanEval and 6.7% Pass@1 and 20.6% Pass@10 on 1-shot MBPP coding benchmarks. Our approach is capable of generating high-quality discrete data in a non-autoregressive fashion, significantly closing the gap between autoregressive models and discrete flow models.

Scores are calculated by giving a weight of 1 for applicable, 0.5 for conditionally applicable, and 0 for incorrect responses. The values are presented as percentages, calculated by the number of responses that satisfy the criteria divided by the total number of responses. The country with the highest percentage is marked in bold, and the second highest is underlined.

Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of diverse regions. That is, information about the food people eat for their birthday celebrations, spices they typically use, musical instruments youngsters play, or the sports they practice in school is common cultural knowledge but uncommon in easily collected online sources, especially for underrepresented cultures.

Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of diverse regions. That is, information about the food people eat for their birthday celebrations, spices they typically use, musical instruments youngsters play, or the sports they practice in school is common cultural knowledge but uncommon in easily collected online sources, especially for underrepresented cultures.

North Korean leader Kim Jong Un supervises the test of suicide drones with artificial intelligence technology, according to local media, at an unknown location, in this photo released by North Korea's official Korean Central News Agency on March 27, 2025. KIM POWER PLAY: North Korean dictator Kim Jong Un oversaw tests of newly developed AI-powered suicide drones and called for their increased production, North Korean state media said Thursday. A photo taken on October 4, 2023 in Manta, near Turin, shows a smartphone and a laptop displaying the logos of the artificial intelligence OpenAI research company and ChatGPT chatbot. SUZANNE'S TWIN: Suzanne Somers passed away two years ago, but her memory lives on, not only through her Hollywood career and businesses, but artificial intelligence too. Her widower, Alan Hamel, worked with an AI company called Hollo to create a "twin" of his late wife.

Despite Flow Matching and diffusion models having emerged as powerful generative paradigms for continuous variables such as images and videos, their application to high-dimensional discrete data, such as language, is still limited. In this work, we present Discrete Flow Matching, a novel discrete flow paradigm designed specifically for generating discrete data. Discrete Flow Matching offers several key contributions: (i) it works with a general family of probability paths interpolating between source and target distributions; (ii) it allows for a generic formula for sampling from these probability paths using learned posteriors such as the probability denoiser (x-prediction) and noise-prediction (ϵ-prediction); (iii) practically, focusing on specific probability paths defined with different schedulers improves generative perplexity compared to previous discrete diffusion and flow models; and (iv) by scaling Discrete Flow Matching models up to 1.7B parameters, we reach 6.7% Pass@1 and 13.4% Pass@10 on HumanEval and 6.7% Pass@1 and 20.6% Pass@10 on 1-shot MBPP coding benchmarks. Our approach is capable of generating high-quality discrete data in a non-autoregressive fashion, significantly closing the gap between autoregressive models and discrete flow models.

During a press conference with Japans Prime Minister Shigeru Ishiba, President Trump said relations with North Korea will happen. North Korean dictator Kim Jong Un has overseen tests of newly developed AI-powered suicide drones and called for their increased production, North Korean state media said Thursday. Photos released from the communist country show Kim inspecting new upgraded reconnaissance drones that are capable of detecting various tactical targets and enemy activities on land and at sea, KCNA state news agency said. Kim said unmanned control and AI capability must be the top priorities in modern arms development. North Korean leader Kim Jong Un has overseen tests of newly developed AI-powered suicide drones and called for their increased production, North Korean state media said.

North Korean leader Kim Jong Un has personally supervised his country's testing of new AI-equipped suicide and reconnaissance drones and called for unmanned aircraft and artificial intelligence to be prioritised in military modernisation plans. State-run Korean Central News Agency (KCNA) said on Thursday that Kim oversaw the testing of "various kinds of reconnaissance and suicide drones" produced by North Korea's Unmanned Aerial Technology Complex. The new North Korean drones are capable of "tracking and monitoring different strategic targets and enemy troop activities on the ground and the sea", while the attack drones will "be used for various tactical attack missions", KCNA said, noting that both drone systems have been equipped with "new artificial intelligence". Kim agreed to expand the production capacity of "unmanned equipment and artificial intelligence" and emphasised the importance of creating a long-term plan for North Korea to promote "the rapid development" of "intelligent drones", which is "the trend of modern warfare". Pictures from the tests, which took place on Tuesday and Wednesday, were said to show attack drones successfully striking ground targets, including military vehicles.

North Korean leader Kim Jong Un supervised the test of suicide drones with artificial intelligence technology and said unmanned control and AI capability must be the top priorities in modern arms development, state media reported on Thursday. Kim inspected new upgraded reconnaissance drones that are capable of detecting various tactical targets and enemy activities on land and at sea, the KCNA state news agency said. "The field of unmanned equipment and artificial intelligence should be top-prioritized and developed in modernizing the armed forces," KCNA quoted Kim as saying.