In open-ended natural-language generation, existing text decoding methods typically struggle to produce text which is both diverse and high-quality. Greedy and beam search are known to suffer from text degeneration and linguistic diversity issues, while temperature, top-k, and nucleus sampling often yield diverse but low-quality outputs. In this work, we present crowd sampling, a family of decoding methods based on Bayesian risk minimization, to address this diversity-quality trade-off. Inspired by the principle of "the wisdom of the crowd," crowd sampling seeks to select a candidate from a pool of candidates that has the least expected risk (i.e., highest expected reward) under a generative model according to a given utility function. Crowd sampling can be seen as a generalization of numerous existing methods, including majority voting, and in practice, it can be used as a drop-in replacement for existing sampling methods. Extensive experiments show that crowd sampling delivers improvements of 3-7 ROUGE and BLEU points across a wide range of tasks, including summarization, data-to-text, translation, and textual style transfer, while achieving new state-of-the-art results on WebNLG and WMT'16.

A surveillance startup in Silicon Valley is being accused of sexism and discrimination after a sales director used the company's facial recognition system to harass female workers. Verkada, which was valued in January at $1.6 billion, equips its office with its own security cameras. Last year, the sales director accessed these cameras to take photos of female workers, then posted them in a Slack channel called #RawVerkadawgz alongside sexually explicit jokes. The incident was first reported by IPVM and independently verified by Vice. Employees told IPVM that a group of men in leadership positions on the sales team, many of whom grew up in Danville and played football together in high school, contributed to a culture of sexism.