In 2023, Noam Chomsky, considered the founder of modern linguistics, wrote that LLMs "learn humanly possible and humanly impossible languages with equal facility." However, in the Mission: Impossible Language Models paper that received a Best Paper award at the 2024 Association of Computational Linguistics (ACL) conference, researchers shared the results of their testing of Chomsky's theory, having discovered that language models actually struggle with learning languages with non-standard characters. Rogers Jeffrey Leo John, CTO of DataChat Inc., a company that he cofounded while working at the University of Wisconsin as a data science researcher, said the Mission: Impossible paper challenged the idea that LLMs can learn impossible languages as effectively as natural ones. "The models [studied for the paper] exhibited clear difficulties in acquiring and processing languages that deviate significantly from natural linguistic structures," said John. "Further, the researchers' findings support the idea that certain linguistic structures are universally preferred or more learnable both by humans and machines, highlighting the importance of natural language patterns in model training. This finding could also explain why LLMs, and even humans, can grasp certain languages easily and not others."

A long-term goal of machine learning research is to build an intelligent dialog agent. Most research in natural language understanding has focused on learning from fixed training sets of labeled data, with supervision either at the word level (tagging, parsing tasks) or sentence level (question answering, machine translation). This kind of supervision is not realistic of how humans learn, where language is both learned by, and used for, communication. In this work, we study dialog-based language learning, where supervision is given naturally and implicitly in the response of the dialog partner during the conversation. We study this setup in two domains: the bAbI dataset of [23] and large-scale question answering from [3]. We evaluate a set of baseline learning strategies on these tasks, and show that a novel model incorporating predictive lookahead is a promising approach for learning from a teacher's response. In particular, a surprising result is that it can learn to answer questions correctly without any reward-based supervision at all.

What would happen if you filled a virtual town with AIs and set them loose? As it turns out, they brush their teeth and are very nice to one another! But this unexciting outcome is good news for the researchers who did it, since they wanted to produce "believable simulacra of human behavior" and got just that. The paper describing the experiment, by Stanford and Google researchers, has not been peer reviewed or accepted for publication anywhere, but it makes for interesting reading nonetheless. The idea was to see if they could apply the latest advances in machine learning models to produce "generative agents" that take in their circumstances and output a realistic action in response. And that's very much what they got.

This particular article is a discussion about a recommendation to a given standard, that of SPARQL 1.1. None of this has been implemented yet, and as such represents more or less the muiings of a writer, rather than established functionality. Lately, I've been spending some time on the Github archives of the SPARQL 1.2 Community site, a group of people who are looking at the next generation of the SPARQL language. One challenge that has come up frequently has been the lack of good mechanisms in SPARQL for handling ordered lists, something that has proven to be a limiting factor in a lot of ways, especially given that most other languages have had the ability of handling lists and dictionaries for decades. As I was going through the archives, an answer occurred to me that comes down to the fact that RDF and SPARQL, while very closely related, are not in fact the same things.

It's an age-old question – what makes someone attractive? We often say things like "beauty is in the eye of the beholder" but while this romantic notion may bring comfort to those dealt a poor hand in life, it also gives the impression that the foundations of attractiveness are elusive and unpredictable. It suggests that what each of us sees as an attractive trait – whether physical or psychological – is so variable that everyone must be looking for something different. While there is variety in what each of us regards as beautiful, cutting through this noise are common and consistent preferences. Psychological traits such as a sense of humor, intelligence, and kindness are generally sought after.

It is an age-old question – what makes someone attractive? We often say things like'beauty is in the eye of the beholder' but while this romantic notion may bring comfort to those dealt a poor hand in life, it also gives the impression that the foundations of attractiveness are elusive and unpredictable. It suggests that what each of us sees as an attractive trait – whether physical or psychological – is so variable that everyone must be looking for something different. Is beauty in the eye of the beholder? Researchers plan to measure dozens of volunteers' characteristics – including humour, intelligence, impulsivity, facial symmetry, strength, and more.

He has more than 20 years of experience in the telecommunications, IT Infrastructure, and security industries. Previously he held positions as general manager data center division and senior vice president core technology at Trend Micro. Before that John was senior director of product management at Lucent Technologies. He has lived and worked in Europe, Asia, and the United States. John graduated with a bachelor of telecommunications engineering degree from Plymouth University, United Kingdom.

With the many, many uses of AI, we're seeing an increase in researchers, scientists, organisations and start-ups of all kinds looking at ways we can leverage this technology for good. Whilst'high-technology' has become synonymous with high wages, and high investment, there are loads of projects out there applying this technology to poverty reduction. Harnessing the power of AI to help the most desperate in our society is a fantastic way to use it. So, how is this being done? Recognising the causes of poverty is key in looking at how to tackle the problems using technologies.

McCarthy, a past president of AAAI and an AAAI Fellow, helped design the foundation of today's internet-based computing and is widely credited with coining the term, artificial intelligence. This remembrance by Edward Feigenbaum, also a past president of AAAI and a professor emeritus of computer science at Stanford University, was delivered at the celebration of John McCarthy's accomplishments, held at Stanford on 25 March 2012. Everyone knew everyone else, and saw them at the few conference panels that were held. At one of those conferences, I met John. We renewed contact upon his rearrival at Stanford, and that was to have major consequences for my professional life.

JOHN GASCHNIC, used to come into my office quite regularly for conversations that helped calibrate my mental compass. John had definite opinions about many subjects-about how N research ought, to be pursued, about high standards of achievement, about the need for first-rate equipment, about important research topics, about personnel mattersin short, about all the sorts of things that concern me. John was persuasive and not easily deflected. Well, actually, he couldn't be deflected at all! Usually, I agreed with John and welcomed the added strength that he gave me. Sometimes, though, I might try to explain that something he was advocating wasn't "realistic."