The Industry Disney's Deal With OpenAI Is So Much Worse Than You Think The $1 billion partnership allows users to create A.I.-generated images of the company's iconic characters. That's not going to end well for anyone. Enter your email to receive alerts for this author. You can manage your newsletter subscriptions at any time. You're already subscribed to the aa_Nitish_Pahwa newsletter.

Elon Musk, the billionaire owner of X, says the advertisers that have stopped spending on the platform due to his endorsement of an antisemitic post can "f----" themselves. "What it's going to do is it's going to kill the company, and the whole world will know the advertisers killed the company," Musk said at the New York Times DealBook conference on Wednesday. The post was the "worst and dumbest I've ever done," said Musk, the chief executive officer of Tesla Inc. Still, if advertisers leave the company, its failure will be their fault, not his -- saying they were trying to "blackmail me with money," he said. "I won't tap dance" to prove trustworthy, he said.

Microsoft's often-overlooked search engine, Bing, is mounting a comeback with ChatGPT, the suddenly ubiquitous chatbot capable of composing song lyrics, writing academic essays and answering all manner of questions. The new version of Bing was released to a limited group of users on Tuesday. The revamped product is part of Microsoft's $13 billion investment in OpenAI, the artificial intelligence lab behind ChatGPT that Microsoft is betting on to stay competitive with its big tech rivals like Google, Apple and Meta. But those companies are also racing to incorporate the new technology into their own software. A day before the unveiling of the new Bing, Google announced that it would soon release an experimental chatbot called Bard for its own search engine, which is much more widely used than Bing.

Foundation models offer an exciting new paradigm for constructing models with out-of-the-box embeddings and a few labeled examples. However, it is not clear how to best apply foundation models without labeled data. A potential approach is to fuse foundation models with weak supervision frameworks, which use weak label sources -- pre-trained models, heuristics, crowd-workers -- to construct pseudolabels. The challenge is building a combination that best exploits the signal available in both foundation models and weak sources. We propose Liger, a combination that uses foundation model embeddings to improve two crucial elements of existing weak supervision techniques. First, we produce finer estimates of weak source quality by partitioning the embedding space and learning per-part source accuracies. Second, we improve source coverage by extending source votes in embedding space. Despite the black-box nature of foundation models, we prove results characterizing how our approach improves performance and show that lift scales with the smoothness of label distributions in embedding space. On six benchmark NLP and video tasks, Liger outperforms vanilla weak supervision by 14.1 points, weakly-supervised kNN and adapters by 11.8 points, and kNN and adapters supervised by traditional hand labels by 7.2 points.

Entity Recognition (ER) is a key component of relation extraction systems and many other natural-language processing applications. Unfortunately, most ER systems are restricted to produce labels from to a small set of entity classes, e.g., person, organization, location or miscellaneous. In order to intelligently understand text and extract a wide range of information, it is useful to more precisely determine the semantic classes of entities mentioned in unstructured text. This paper defines a fine-grained set of 112 tags, formulates the tagging problem as multi-class, multi-label classification, describes an unsupervised method for collecting training data, and presents the FIGER implementation. Experiments show that the system accurately predicts the tags for entities. Moreover, it provides useful information for a relation extraction system, increasing the F1 score by 93%. We make FIGER and its data available as a resource for future work.