A former ByteDance intern, who was reportedly fired for professional misconduct including sabotaging the work of colleagues, was announced this week as the winner of one of the most prestigious annual awards for AI research. Qiu Tian, whose LinkedIn and Google Scholar pages list him as a master's student in computer science at Peking University, was among two papers selected Tuesday for the main “Best Paper Award” in Neural Information Processing Systems (NEURIPS). Is the first author of one of the. conference, biggest gathering of machine learning researchers in the world.
paperEntitled “Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction,” a new method for creating AI-generated images is presented by Tian and four co-authors – all from ByteDance or Peking University. Affiliated with – claims it is faster and more efficient than it is. Predecessor. “The overall quality of the paper presentation, experimental verification, and insight (scaling law) give compelling reasons to experiment with this model,” the NeuReps Best Paper Award committee wrote in a statement. statement,
The committee's decision to honor Tian, which ByteDance reportedly said filed a lawsuit against Last month it became the center of widespread discussion online about how NeuroIPS is run and topped up, claiming deliberate sabotage of other company research projects, leading to losses of more than $1 million. AI researcher Evaluate the work of your coworkers. This news led to details of the scam that had been circulating on Chinese social media for several weeks finally spreading to the English-language Internet.
“NeurIPS gave the best paper award to a highly problematic task (not the first time, by the way),” Abeba Birhane, Head newly formed AI Accountability Lab at Trinity College, wrote On BlueSky. “You would think that a conference that prides itself on upholding the highest scientific and ethical standards [do] “Do your due diligence before giving an award to a newspaper that directly contradicts their values.”
A spokesperson for NeuIPS stressed that the honor was given to the newspaper, not Tian himself. He directed WIRED to a portion of the awards committee statement Explaining how the conference evaluates paper submissions. It reads, “The search committees considered all accepted NeuroIPS papers equally, and independently based on the scientific merit of the papers, without separately considering authorship or other factors, in keeping with the NeuroIPS blind review process.” Took decisions from.”
Bluesky, Birhen and other AI researchers linked to an anonymous GitHub blog post It also circulated on HackerNews, Reddit, and other platforms in recent days, urging the academic AI community to reconsider awarding best paper honors to Tian due to his “serious misconduct”, stating that “fundamentally “undermines the core values of integrity and trust.” Which our academic community creates.”