A paper co-authored by former ByteDance intern Tian Keyu reportedly won a Best Paper award at NeurIPS, one of the most prestigious annual conferences in the field of AI. The award comes after Tian was sued by ByteDance for allegedly disrupting big-model training. The paper, titled “Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction,” explores a unique method for generating images with faster speed, higher quality, and scalability. It was published in collaboration with Beijing University by ByteDance’s commercialization technology department and completed by five authors, with Tian listed as the lead author. The award is expected to be presented at the NeurIPS conference in Vancouver from Dec. 10 to 15. In November, ByteDance filed a lawsuit against Tian, accusing him of tampering with code to disrupt the company’s internal model training. The case has been officially accepted by the Haidian District People’s Court in Beijing. ByteDance is seeking RMB 8 million ($1.1 million) in damages and RMB 20,000 ($275) in reasonable expenses, along with a public apology. [36Kr, in Chinese]
Related
