Solving math word problems is the task that analyses the relation of quantities and requires an accurate understanding of contextual natural language information. Recent studies show that current models rely on shallow heuristics to predict solutions and could be easily misled by small textual perturbations. To address this problem, we propose a Textual Enhanced Contrastive Learning framework, which enforces the models to distinguish semantically similar examples while holding different mathematical logic. We adopt a self-supervised manner strategy to enrich examples with subtle textual variance by textual reordering or problem re-construction. We then retrieve the hardest to differentiate samples from both equation and textual perspectives and guide the model to learn their representations. Experimental results show that our method achieves state-of-the-art on both widely used benchmark datasets and also exquisitely designed challenge datasets in English and Chinese. \footnote{Our code and data is available at \url{https://github.com/yiyunya/Textual_CL_MWP}

Examining a 20th-century Scandinavian legal theoretical tradition, we can extract an ontological naturalistic, a logical empiristic, and a modern idealistic rationale. We introduce the mathematical syntactic figure present in the `logical empiricism' in a contemporary mathematical logic. A new formal framework for describing explicit purchase statutes (Sweden) is gradually developed and subsequently proposed. This new framework is based on a many-sorted first-order logic (MFOL) approach, where the semantics are grounded in concrete `physical' objects and situations with a legal relevance. Specifically, we present a concrete formal syntactic translation of one of the central statutes of Swedish legislation for the purchase of immovable property. Additionally, we discuss the potential implications that a subsequent development of such formalisations would have for constructing artificial agents (e.g., software) that can be used as `co-creative' legal assistance for solving highly complex legal issues concerning the transfer of property, among others.

This note is a sketch of how the concept of robopsychology and robophilosophy could be reinterpreted and repositioned in the spirit of the original vocation of psychology and philosophy. The notion of the robopsychology as a fictional science and a fictional occupation was introduced by Asimov in the middle of the last century. The robophilosophy, on the other hand, is only a few years old today. But at this moment, none of these new emerging disciplines focus on the fundamental and overall issues of the development of artificial general intelligence. Instead, they focus only on issues that, although are extremely important, play a complementary role, such as moral or ethical ones, rather than the big questions of life. We try to outline a conception in which the robophilosophy and robopsychology will be able to play a similar leading rule in the progress of artificial intelligence than the philosophy and psychology have done in the progress of human intelligence. To facilitate this, we outline the idea of a visual artificial language and interactive theorem prover-based computer application called Prime Convo Assistant. The question to be decided in the future is whether we can develop such an application. And if so, can we build a computer game on it, or even an esport game? It may be an interesting question in order for this game will be able to transform human thinking on the widest possible social scale and will be able to develop a standard mathematical logic-based communication channel between human and machine intelligence.

Many attempts to develop artificial intelligence are powered by powerful systems of mathematical logic. They tend to produce results that make logical sense to a computer program -- but the result is not very human. In our work building therapy chatbots, we have found using a different kind of logic -- one first formalised by the Greek philosopher Aristotle more than 2,000 years ago -- can produce results that are more fallible, but also much more like real people. The underpinning science of our chatbots is formal logic. Modern formal logic has its basis in mathematics -- but that wasn't always the case.

If today's college students could find a way to get their hands on a copy of Facebook's latest neural network, they could cheat all the way through Calc 3. They could even solve the differential equation pictured above in under 30 seconds. Okay, so maybe this isn't going to be a replacement for Wolfram Alpha anytime soon, but Facebook really did build a neural net that can complete complex mathematical problems for the first time, rather than the plain old arithmetic in which these AI models usually wheel and deal. The work represents a huge leap forward in computers' abilities to understand mathematical logic. The research is outlined in a new paper, "Deep Learning for Symbolic Mathematics," published in arXiv, a repository of scientific research in areas like math, computer science, and physics, run by Cornell University.

During the twentieth century, discoveries in mathematical logic revolutionized our understanding of the very foundations of mathematics. In 1931, the logician Kurt Gödel showed that, in any system of axioms that is expressive enough to model arithmetic, some true statements will be unprovable1. And in the following decades, it was demonstrated that the continuum hypothesis -- which states that no set of distinct objects has a size larger than that of the integers but smaller than that of the real numbers -- can be neither proved nor refuted using the standard axioms of mathematics2–4. They identify a machine-learning problem whose fate depends on the continuum hypothesis, leaving its resolution forever beyond reach. Machine learning is concerned with the design and analysis of algorithms that can learn and improve their performance as they are exposed to data.

This is still not directly definable, although we still know of human abilities that even the the best present programs on the fastest computers have not been able to emulate, such as playing master-level go and learning science from the Internet. Basic researchers in AI should measure their work as to the extent to which it advances this goal. AI research should not be dominated by near-term applications. DARPA should recall the extent to which its applied goals were benefitted by basic research. NSF should not let itself be seduced by impatience.

The history of computers is often told as a history of objects, from the abacus to the Babbage engine up through the code-breaking machines of World War II. In fact, it is better understood as a history of ideas, mainly ideas that emerged from mathematical logic, an obscure and cult-like discipline that first developed in the 19th century. Mathematical logic was pioneered by philosopher-mathematicians, most notably George Boole and Gottlob Frege, who were themselves inspired by Leibniz's dream of a universal "concept language," and the ancient logical system of Aristotle. Dixon goes on to describe the creation of Boolean logic (which has only two values: TRUE and FALSE, represented as 1 and 0 respectively), and the insight by Claude E. Shannon that those two variables could be represented by a circuit, which itself has only two states: open and closed.1 Dixon writes: Another way to characterize Shannon's achievement is that he was first to distinguish between the logical and the physical layer of computers. Dixon is being modest: the distinction may be obvious to computer scientists, but it is precisely the clear articulation of said distinction that undergirds Dixon's remarkable essay; obviously "computers" as popularly conceptualized were not invented by Aristotle, but he created the means by which they would work (or, more accurately, set humanity down that path).

If ability to perform complex calculations were a sufficient criterion, then even a conventional digital computor could lay claim to more intelligence than any of usand perhaps we had better let it make away with the word and be done with it.

The long-term goal of AI is human-level AI. This is still not directly definable, although we still know of human abilities that even the the best present programs on the fastest computers have not been able to emulate, such as playing master-level go and learning science from the Internet. Basic researchers in AI should measure their work as to the extent to which it advances this goal.