Shinji Watanabe, an associate professor in Carnegie Mellon University's School of Computer Science, has been named a fellow of the International Speech Communication Association (ISCA) "for wide-ranging, fundamental contributions to research and leadership in speech recognition technologies." Founded in 2007, the ISCA Fellows Program recognizes and honors outstanding ISCA members who have made significant contributions to the science and technology of speech communication. Fellows are nominated by association members and selected by a committee of their peers. Since its inception, the program has recognized nearly 100 fellows from countries around the globe. Watanabe, who is part of CMU's Language Technologies Institute, studies automatic speech recognition, speech enhancement, spoken language understanding, and machine learning for speech and language processing.

We've collected some of the articles, opinion pieces, videos and resources relating to large language models. Some of these links also cover other generative models. We will periodically update this list to add any further resources of interest.

Eden Prairie, Minnesota, April 05, 2023 (GLOBE NEWSWIRE) -- Starkey is proud to announce its all-new Genesis AI hearing aids have received a Red Dot Award: Product Design 2023, marking the second award won by the completely redesigned hearing technology, just weeks after its launch. This is the seventh year Starkey has won this award, which is one of the most renowned international product competitions in the world. The annual awards program recognizes the year's best products that are aesthetically appealing, functional, innovative, and most importantly, have outstanding design. "At Starkey, product development begins by pushing the edge of what's possible," said President and CEO, Brandon Sawalich. "Five years ago, we set out to make the impossible possible when we began to imagine our next-level product offering. Receiving this honor is a tribute to the amount of research and development we devoted to producing our all-new hearing technology, which is making a real impact on reducing the stigma around hearing aids."

Originally published on Towards AI. So, what exactly is prompting? In a nutshell, it's communicating with AI through written text or other content, such as images. This new skill is what Sander calls "artificial social intelligence." But why bother learning prompting?

The ChatGPT revolution opens the door to a four-day work week by providing a major productivity boost for swathes of jobs, according to a Nobel Prize-winning labor economist. Christopher Pissarides--a professor at the London School of Economics who specializes in the impact of automation on work--said the labor market can adapt quickly enough to artificial intelligence-backed chatbots. His remark may tamp down concerns that rapid advances in technology could bring mass job losses. "I'm very optimistic that we could increase productivity," he said in an interview at a conference in Glasgow. "We could increase our well-being generally from work and we could take off more leisure. We could move to a four-day week easily."

Most recent self-supervised learning methods are pre-trained on the well-curated ImageNet-1K dataset. In this work, given the excellent scalability of web data, we consider self-supervised pre-training on noisy web sourced image-text paired data. First, we conduct a benchmark study of representative self-supervised pre-training methods on large-scale web data in a like-for-like setting. We compare a range of methods, including single-modal ones that use masked training objectives and multi-modal ones that use image-text constrastive training. We observe that existing multi-modal methods do not outperform their single-modal counterparts on vision transfer learning tasks. We derive an information-theoretical view to explain these benchmark results, which provides insight into how to design a novel vision learner. Inspired by this insight, we present a new visual representation pre-training method, MUlti-modal Generator~(MUG), that learns from scalable web sourced image-text data. MUG achieves state-of-the-art transfer performance on a variety of tasks and demonstrates promising scaling properties. Pre-trained models and code will be made public upon acceptance.

We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate its capabilities on numerous tasks and find that its zero-shot performance is impressive -- often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at https://segment-anything.com to foster research into foundation models for computer vision.

Linguistic pragmatics state that a conversation's underlying speech acts can constrain the type of response which is appropriate at each turn in the conversation. When generating dialogue responses, neural dialogue agents struggle to produce diverse responses. Currently, dialogue diversity is assessed using automatic metrics, but the underlying speech acts do not inform these metrics. To remedy this, we propose the notion of Pragmatically Appropriate Diversity, defined as the extent to which a conversation creates and constrains the creation of multiple diverse responses. Using a human-created multi-response dataset, we find significant support for the hypothesis that speech acts provide a signal for the diversity of the set of next responses. Building on this result, we propose a new human evaluation task where creative writers predict the extent to which conversations inspire the creation of multiple diverse responses. Our studies find that writers' judgments align with the Pragmatically Appropriate Diversity of conversations. Our work suggests that expectations for diversity metric scores should vary depending on the speech act.

Klaus Teuber, who created the board game sensation Settlers of Catan, has passed away at the age of 70. According to the official Catan website, Teuber died on April 1st after a "brief and severe illness." He is survived by his wife Claudia and their two sons, Guido and Benny, all of which are involved with publisher Catan Studios. It is hard to do justice to how important Catan was (and is) to the board game space. Teuber invented the multiplayer resource-trading game in 1995 and is one of the first European board games to achieve massive success throughout the rest of the world.

Transformer networks have revolutionized NLP representation learning since they were introduced. Though a great effort has been made to explain the representation in transformers, it is widely recognized that our understanding is not sufficient. One important reason is that there lack enough visualization tools for detailed analysis. In this paper, we propose to use dictionary learning to open up these "black boxes" as linear superpositions of transformer factors. Through visualization, we demonstrate the hierarchical semantic structures captured by the transformer factors, e.g., word-level polysemy disambiguation, sentence-level pattern formation, and long-range dependency. While some of these patterns confirm the conventional prior linguistic knowledge, the rest are relatively unexpected, which may provide new insights. We hope this visualization tool can bring further knowledge and a better understanding of how transformer networks work. The code is available at https://github.com/zeyuyun1/TransformerVis