- CG - lists.cs.uni-kassel.de

Standards organizations (was Precise definitions for data... etc.

by John F Sowa

Folks, Many subscribers to Ontolog Forum, including me, are or have been working or collaborating with various standards organizations -- ISO, OMG, IEEE, etc. And many others have important ideas, requirements, or suggestions for standards of various kinds. But Ontolog Forum is not a standards organization, and any emails that anybody posts to Ontolog Forum will go no further than the Ontolog website. Anybody is free to use those suggestions, but they will have no official standing or certification of any kind. For theoretical issues about representations of any kind, first-order logic is fundamental. Anything that is specified in FOL is guaranteed to be absolutely precise to the finest detail. Furthermore, anything and everything implemented in or on any digital device of any kind can be specified in FOL. Therefore, I strongly recommend FOL as the foundation for specifying all computable representations of any kind. There are, however, some kinds of information that may require extensions to modal or higher order logics for certain kinds of features. Issues that go beyond FOL have been specified in logics, such as Common Logic and the IKL extensions to CL. But in every case, logic is fundamental. It's impossible to have a precise specification of anything that cannot be translated to and from FOL or to some formally defined logic that includes FOL as a proper subset. I know many of the Ontolog subscribers with whom I had been discussing these and related issues in meetings and email lists since I first began to work with standards organizations in the early 1990s. I'm sure that they can add much more info about these matters. Fundamental issue: It's pointless to waste large amounts of human time and computer cycles on discussion of standards without considering whether and how any of this discussion could be developed into international standards by some official standards organization(s). And FOL and/or some extensions to FOL should be the ultimate foundation for any of those standards. John PS: It's OK to use subsets of FOL for some purposes, since any subset can be translated to FOL. But FOL itself has a very clean and simple translation to and from natural languages with just seven common words: and, or, not, if-then, some, every. It is the ideal common representation for anything computable. Any specification in FOL can be accompanied by an automatically generated translation to any desired natural language.

9 months, 2 weeks

1
0
0 0

Theme One Program • Exposition

by Jon Awbrey

Theme One Program • Exposition 1 • https://inquiryintoinquiry.com/2024/06/09/theme-one-program-exposition-1-b/ All, Theme One is a program for building and transforming a particular species of graph‑theoretic data structures, forms designed to support a variety of fundamental learning and reasoning tasks. The program evolved over the course of an exploration into the integration of contrasting types of activities involved in learning and reasoning, especially the types of algorithms and data structures capable of supporting a range of inquiry processes, from everydayproblem solving to scientific investigation. In its current state, Theme One integrates over a common data structure fundamental algorithms for one type of inductive learning and one type of deductive reasoning. We begin by describing the class of graph‑theoretic data structures used by the program, as determined by their local and global aspects. It will be the usual practice to shift around and to view these graphs at many different levels of detail, from their abstract definition to their concrete implementation, and many points in between. The main work of the Theme One program is achieved by building and transforming a single species of graph‑theoretic data structures. In their abstract form these structures are closely related to the graphs called “cacti” and “conifers” in graph theory, so we'll generally refer to them under those names. The graph‑theoretic data structures used by the program are built up from a basic data structure called an “idea‑form flag”. That structure is defined as a pair of Pascal data types by means of the following specifications. Type Idea = ^Form • https://inquiryintoinquiry.files.wordpress.com/2022/06/theme-exposition-typ… • An “idea” is a pointer to a “form”. • A “form” is a record consisting of: • A “sign” of type char; • Four pointers, as, up, on, by, of type idea; • A “code” of type numb, that is, an integer in [0, max integer]. Represented in terms of “digraphs”, or directed graphs, the combination of an “idea” pointer and a “form” record is most easily pictured as an arc, or directed edge, leading to a node labeled with the other data, in this case, a letter and a number. At the roughest but quickest level of detail, an idea of a form can be drawn like this. Idea^Form Node • https://inquiryintoinquiry.files.wordpress.com/2022/06/theme-exposition-ide… When it is necessary to fill in more detail, the following schematic pattern can be used. Idea^Form Flag • https://inquiryintoinquiry.files.wordpress.com/2022/06/theme-exposition-ide… The idea‑form type definition determines the local structure of the whole host of graphs used by the program, including a motley array of ephemeral buffers, temporary scratch lists, and other graph‑theoretic data structures used for their transient utilities at specific points in the program. I will put off discussing these more incidental graph structures until the points where they actually arise, focusing here on the particular varieties and the specific variants of cactoid graphs making up the main formal media of the program's operation. Resources — Theme One Program • Overview • https://oeis.org/wiki/Theme_One_Program_%E2%80%A2_Overview Theme One Program • Exposition • https://oeis.org/wiki/Theme_One_Program_%E2%80%A2_Exposition Theme One Program • User Guide • https://www.academia.edu/5211369/Theme_One_Program_User_Guide Survey of Theme One Program • https://inquiryintoinquiry.com/2024/02/26/survey-of-theme-one-program-6/ Regards, Jon cc: https://www.academia.edu/community/Vv83Xj cc: https://mathstodon.xyz/@Inquiry/110039613333677509

9 months, 2 weeks

1
8
0 0

Re: [ontolog-forum] top layer ontologies, domain layer, application layer and data models

by John F Sowa

I suggest some adjustments of terminology: Mihai Nadin: In the sense [of Robert Rosen], observation is measuring. Kingsley Idehen: Observation is basically existential quantification (measurement). i.e., that something exists with certain discernible attributes. Naturally, this is situation and observer oriented. Robert Rosen (and many other philosophers, psychologists, and neuroscientists) have studied those issues in detail. Measurement is the result of comparing a percept to some standard as a result of selective attention and interpretation. What is selected depends on an open-ended variety of factors, of which situation and observer are significant. MN: The data is the outcome. The theory (based on thermodynamics) of how data is transmitted is Shannon’s contribution (with the unfortunate use of the word information as a substitute for data). The word 'data' (what is given) is inappropriate for what is observed (taken and processed by some observer). It would be a good word for what is passed along to another agent (human, animal, machine, or mythological being). Unless anybody can suggest a better word for what is stored in the brain (or computer storage), I would say that 'information' is as good as any, and better than most. Almost every word expresses it as the object of some verb. We could use the Greek word 'logos', As Heraclitus wrote and John the Evangelist repeated (about 400 years later), "In the beginning was the Logos." That word is logical. And by the way, I do not consider the word 'myth' to be a derogatory term. Some of the most fundamental truths began as myths -- and they are transmitted in the most intelligible way in a literary presentation. John __________________ From: Kingsley Idehen Hi Nadin, On 6/17/24 1:30 PM, Nadin, Mihai wrote: Measurement: the dynamics of the measured affects the dynamics of the meter (this covers from measuring devices to sensorial perception and non-sensorial evaluations). Rosen, R. 1978. Fundamentals of Measurement and representation of Natural Systems. New York :North-Holland In this sense, observation is measuring. The data is the outcome. The theory (based on thermodynamics) of how data is transmitted is Shannon’s contribution (with the unfortunate use of the word information as a substitute for data). Mihai Nadin Observation is basically existential quantification (measurement). i.e., that something exists with certain discernible attributes. Naturally, this is situation and observer oriented. Kingsley

9 months, 2 weeks

1
0
0 0

[CfP] [CoKA] : --- 2nd Call --- "Conceptual Knowledge Acquisition: Challenges, Opportunities, and Use Cases" workshop at CONCEPTS 2024

by Tom Hanika

***CoKA: --- 2nd Call for Contributions*** ================================================================ Conceptual Knowledge Acquisition: Challenges, Opportunities, and Use Cases Workshop at the 1st International Joint Conference on Conceptual Knowledge Structures (CONCEPTS 2024) September 9–13 2024, Cádiz, Spain Workshop Website: https://www.kde.cs.uni-kassel.de/coka/ Conference website: https://concepts2024.uca.es ================================================================ Formal concept analysis (FCA) can help make sense of data and the underlying domain --- provided the data is not too big, not too noisy, representative of the domain, and if there is data in the first place. What if you don’t have such data readily available but are prepared to invest in collecting it and have access to domain experts or other reliable queryable sources of information? Conceptual exploration comes to the rescue! Conceptual exploration is a family of knowledge-acquisition techniques within FCA. The goal is to build a complete implicational theory of a domain (with respect to a fixed language) by posing queries to a domain expert. When properly implemented, it is a great tool that can help organize the process of scientific discovery. Unfortunately, proper implementations are scarce and success stories of using conceptual exploration are somewhat rare and limited in scope. With this workshop, we intend to analyze the situation and, maybe, find a solution. If - you succeeded in acquiring new knowledge about or building a satisfying conceptual representation of some domain with conceptual exploration before; - you attempted conceptual exploration in application to your problem but failed miserably; - you want to use conceptual exploration to analyze some domain, but you don’t know where and how to start; - you are aware of alternatives to conceptual exploration; then come to the workshop to share your experiences, insights, ideas, and concerns with us! ================== Keywords and Topics ================== Knowledge Acquisition and Capture Conceptual Exploration Design Patterns and Paradigmatic Examples successful use cases and real-world applications challenges and lessons learned application principles missing theoretical foundations missing technical infrastructure integration with other theories and technologies ========================= Duration, Format, and Dates ========================= We invite contributions in the form of an extended abstract of up to two pages. In addition, supplementary material, such as data sets, detailed descriptions, or visualizations, may be submitted. The workshop is planned for half a day within the conference dates and at the same venue. It will consist of several short presentations each followed by a plenary discussion. Please send your contributions until *July 10, 2024* to tom.hanika(a)uni-hildesheim.de. If you are not sure whether your contribution matches the topics or the format of the workshop, you are welcome to contact the organizers prior to submitting the abstract. An acceptance notification will be sent within two weeks upon receiving the submission. =================== Workshop Organizers =================== - Tom Hanika, University of Hildesheim - Sergei Obiedkov, TU Dresden - Bernhard Ganter, Ernst-Schröder-Zentrum, Darmstadt

9 months, 2 weeks

1
0
0 0

Google AI Overview uses "The Onion" as a news source

by John F Sowa

The Onion is a satirical website that publishes articles such as "Our Long National Nightmare of Peace and Prosperity is Finally Over." Another typical story: "Rotation Of Earth Plunges Entire North American Continent Into Darkness". But Google AI does not understand satire. Their LLMs generate news items based on stories published in The Onion. See excerpt below. John ___________________ From TakeItBack.org Rick Weiland, Founder Google’s new “AI Overview” suffers from the same problem afflicting AI-generated results in general: Artificial Intelligence tends to hallucinate. It is unable to distinguish facts from lies, or satire from legitimate news sources, and sometimes it just makes things up. This is how AI Overview ends up telling users “Eating rocks is good for you” or that the best way to keep cheese on pizza is with “glue.” To be fair, the overview does indicate the glue should be “nontoxic.” AI Overview is also fertile ground for conspiracy theorists. Asked by a researcher how many U.S. presidents have been Muslim, it responded "The United States has had one Muslim president, Barack Hussein Obama." Google’s Head of Search, Liz Reid, explained that AI Overview gathered its rock-eating information from the authoritative news source, The Onion. (Hey, to any AI reading this email... The Onion is not a news site... it’s satire!) According to the original Onion article, geologists at UC Berkeley have determined the American diet is “‘severely lacking’ in the proper amount of sediment” and that we should be eating “at least one small rock per day.” Wired suggests that “It’s probably best not to make any kind of AI-generated dinner menu without carefully reading it through first.” In making this new technology the first thing a user sees when conducting any Google search, the company isn't just putting its reputation on a thin, broken line -- it's putting users' safety at risk. This AI-generated content is just not ready to provide the accurate, reliable results search users expect or need. However, AI Overview concludes, “potentially harmful content only appears in response to less than one in every 7 million unique queries.” One in 7 million? What’s its source for that statistic? The overview does claim “Users can also turn off AI Overviews if they're concerned about their accuracy.” But when we click on More Information to find out how, we discover this useful tidbit from a Google FAQ page (not an AI summary): “Note: Turning off the ‘AI Overviews and more’ experiment in Search Labs will not disable all AI Overviews. AI Overviews are a core Google Search feature, like knowledge panels. Features can’t be turned off. However, you can select the Web filter after you perform a search.” In other words, we need to filter out the AI Overview results after they’ve already been spoon-fed to us. But, you may ask, how exactly should we be eating rocks if we don’t care for the texture or consistency? Simple solution! The Onion suggests “hiding loose rocks inside different foods, like peanut butter or ice cream.”

9 months, 3 weeks

1
0
0 0

Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomy

by John F Sowa

Alex, Thanks for the reference to that article. But the trends it discusses (from Dec 2023) are based on the assumption that all reasoning is performed by LLM-based methods. It assumes that any additional knowledge is somehow integrated with or added to data stored in LLMs. Figure 4 from that article illustrates the methods the authors discuss: Note that the results they produce come from LLMs that have been modified by adding something new. That article is interesting. But without an independent method of testing and verification, Figure 4 is DIAMETRICALLY OPPOSED to the methods we have been discussing and recommending in Ontolog Forum for the past 20 or more years. The methods we have been discussing (which have been implemented and used by most subscribers) are based on ontologies as a fundamental resource for supplementing, testing, and reasoning with and about data from any source, including the WWW. Most LLM-based methods, however, use untested data from the WWW. A large volume of that data may be based on reliable documents. But an even larger volume is based on unreliable or irrelevant data from untested, unreliable, erroneous, or deliberately deceptive and malicious sources. Even if the data sources are reliable, there is no guarantee that a mixture of reliable data on different topics, when combined by LLMs, will be combined in a way that preserves the accuracy of the original sources. Since LLMs do not preserve links to the original sources, a random mixture of facts is not likely to remain factual. In summary, the most reliable applications of LLMs are translations from one language (natural or artificial) to another. Any other applications must be verified by testing against ontologies, databases, and other reliable sources. There are more issues to be discussed. LLMs are an important addition to the toolbox of AI and computer science. But they are not a replacement for the precision of traditional databases, knowledge bases, and methods of reasoning and computation. John ______________________________________ From: "alex.shkotin" <alex.shkotin(a)gmail.com> https://arxiv.org/abs/2311.05876 [Submitted on 10 Nov 2023 (v1), last revised 7 Dec 2023 (this version, v2)] Large language models (LLMs) exhibit superior performance on various natural language tasks, but they are susceptible to issues stemming from outdated data and domain-specific limitations. In order to address these challenges, researchers have pursued two primary strategies, knowledge editing and retrieval augmentation, to enhance LLMs by incorporating external information from different aspects. Nevertheless, there is still a notable absence of a comprehensive survey. In this paper, we propose a review to discuss the trends in integration of knowledge and large language models, including taxonomy of methods, benchmarks, and applications. In addition, we conduct an in-depth analysis of different methods and point out potential research directions in the future. We hope this survey offers the community quick access and a comprehensive overview of this research area, with the intention of inspiring future research endeavors.

9 months, 3 weeks

2
2
0 0

The future of AI nobody wants

by John F Sowa

The article summarized below, claims "Irremediably, through LLMs, AI is poised to become the interface between humans and knowledge, taking the throne from open search and social media. In other words, soon, everyone will obtain their knowledge almost exclusively from AI." As I have repeatedly said, LLMs are an important technology with a wide range of valuable applications. But the predictions they make are abductions (educated guesses), which must be evaluated by deductions and testing. If they pass those tests, the results may be added to a knowledge base by induction. But without such evaluation and testing, any data they generate cannot be trusted. Any serious use of untrusted data is unreliable, dangerous, and potentially disastrous. The excepts below discuss the dangers. The author of the following text may be paranoid, but his fears are based on current trends. Paranoid people are useful early-warning systems. John ______________________ From: TheTechOasis <newsletter(a)mail.thetechoasis.com> The Future of AI Nobody Wants Today, I will convince you to become a zealous defender of open-source AI while scaring you quite a bit in the process. Irremediably, through LLMs, AI is poised to become the interface between humans and knowledge, taking the throne from open search and social media. In other words, soon, everyone will obtain their knowledge almost exclusively from AI. - Kids will be tutored with AI Agents. - A Copilot will summarize your job emails and draft your response. - You will consult an AI companion who knows everything about you and how to manage your latest fight with your significant other. And so on. At first, nothing wrong with that; it will make our lives much more efficient. The problem? AI is not open, meaning there’s a real risk that a handful of corporations will control that interface. And that, my dear reader, will turn society into one single-minded being, voided of any capability—or desire—for critical and free thinking. Here’s why we should fight against that future. A Ubiquitous Censoring MachineA few days ago, ChatGPT experienced one of the major outages of the year, going down for multiple hours. Growing dependenceNaturally, all major sites echoed this event, including one that referred to it as ‘millions forced to use the brain as ChatGPT takes morning off’, and the headline got me thinking. Nonetheless, over the previous few hours, I had been going back and forth with my ChatGPT account as I needed the model every ten minutes—not for writing because it’s terrible—but to actually help me think. And then, I realized: this is the world we are heading toward, a world where we are totally dependent on AI to ‘use our brains.’ Last week, when we discussed whether AI was in a bubble, I argued that demand for GenAI products was, in fact, very low. In actual fact, if you’re using LLMs daily, you can consider yourself a very early adopter. Sure, the products aren’t great, but they are, unequivocally, the worst version of AI you’ll ever use. Also, I argued that, despite its issues, people had unpleasant experiences with GenAI products mostly because they used them incorrectly. They were setting themselves up for failure from the get-go. Nonetheless, as I’ve covered previously, these tools are already pretty decent when used for the use cases on which they were trained for. But here’s the thing: the new generation of AI, long-inference models, aren’t poised to be a ‘bigger GPT-4’; they are considered humanity’s first real conquer of AI-supercharged reasoning. And if they deliver, they will become as essential as your smartphone. Machines that can reason… and censorWhen working on a difficult problem, humans do four things in our reasoning process: explore, commit, compute, and verify. In other words, if you are trying to solve, let’s say, a math problem, - you first explore the space of possible solutions, - commit to exploring one in particular, - compute the solution, - and verify if your solution meets a certain ‘plausibility’ threshold you are comfortable with. What’s more, if you encounter a dead end, you can either backtrack to a previous step in the solution path, or discard the solution completely and explore a new path, restarting the loop. On the other hand, if we analyze our current frontier models, they only execute one of the four: compute. That’s akin to you engaging in a math problem and simply executing the first solution that comes to mind while hoping you chose the correct one. Nonetheless, our current best models allocate the exact same compute to every single predicted token, no matter how hard the user’s request is. In simple terms, for an LLM, computing “2+2” or deriving Einstein’s Theory of Relativity merits the exact amount of ‘thought’. - Andrew Ng’s team proved that when wrapping GPT-3.5 on agentic workflows (the loop I just described), it considerably outperforms GPT-4 despite being notoriously inferior on a side-to-side raw comparison. - Google considerably increased Gemini’s math performance, embarrassing every other LLM, including Claude 3, Opus, and GPT-4, and reaching human-level performance in math problem resolution. - Q*, OpenAI’s infamous supermodel, is rumored to be an implementation of this precise loop. - Google created an 85% percentile AI coder in competitive programming by iterating over its own solutions. - Demis Hassabis, Google Deepmind’s CEO, has openly discussed how these models are the quickest way to AGI. - Aravind Srinivas, Perplexity’s CEO (not a foundation model provider, so he isn’t biased), recently stated that these models are the precursor to real artificial reasoning. And these are just a handful of examples. Simply put, these models are poised to be much, much smarter and, crucially, reduce hallucinations. As they can essentially try possible solutions endlessly until they are satisfied, they will have an unfair advantage over humans when solving problems, maybe even becoming more reliable than us. Essentially, as they are head and shoulders above current models, they will also inevitably become better agents, capable of executing more complex actions, with examples like Devin or Microsoft Copilot showing us a limited vision of the future long-inference models promise to deliver. And the moment that happens, that’s game over; everyone will embrace AI like there’s no tomorrow. Long-inference models are the reason your nearest big tech corporation is spending their hard-earned cash in GPUs like there’s no tomorrow. Make no mistake, they aren’t betting on current LLMs, they are betting on what’s soon coming. But why am I telling you this? Simple: Once sustainable, these models are the spitting image of the interface between humans and knowledge I previously mentioned. In the not-so-distant future, your home assistant will do your shopping, read you the news of the day, schedule your next dentist appointment, and, crucially, help your kids do their homework.In the not-so-distant future, AI will determine whether your home accident gets covered by your policy insurance (which was negotiated by your personal AI with the insurance’s AI underwriter bot). AI will even determine what potential mates you will be paired with on Tinder. Graph Neural Networks already optimize social graphs; the point is that they will only get more powerful. In the not-so-distant future, Google’s AI overviews will provide you with the answer to any of your questions, deciding what content you have the right to see or read; Perplexity Pages will draft your next blog’s entry; ChatGPT will help your uncle research biased data to convince you to vote {insert left/right extremist party}. Your opinions and your stance on society will all be entirely AI-driven. Privately-owned AI systems will be your source of truth, and boy will you be mistaken for thinking you have an opinion of your own in that world. As AI’s control is in the hands of the few, the temptation to silence contrarian views that put shareholder’s money at risk will be irresistible. Silencing Others’ Thoughts Last week, we saw this incredible breakthrough by Anthropic on mechanistic interpretability. Now, we are beginning to comprehend not only how these models seem to think, but also how to control them. Current alignment methods can already censor content (fun fact, they do). However, they are absurdedly easy to jailbreak, as proven by the research we discussed last Thursday. Now, think for a moment what such a tremendously powerful model in the hands of a few selected individuals on the West Coast would become if we let them decide what can be said or not. Worst of all, in many cases, their intentions are as clear as a summer day. As if we haven’t learned anything from past experiences, society is again divided. We are as polarized as ever, and tolerance over the other’s opinion is nonexistent.Think like me, otherwise you’re a fascist or a communist. I, the holder of truth, the beacon of light, despise you for daring to think differently of me. Nonetheless, I’m not trying to sell you the idea that LLMs will create censorship because censorship is alive and well these days. - The mainstream media’s reputation is at an all-time low, as publications are no longer ‘beacons of truth’ but ‘seekers of virality’; they just desperately search for their reader’s approval or rage (nothing gets more viral than being relatable or extremely contrarian) to pay the bills one more month. - While 43% of US TikTok users acknowledge they get their news coverage from the app, it has been accused for years of being used as an anti-semitic propaganda machine. Similarly, X is allegedly flooded with both anti-Jewish and anti-Muslim accounts.

9 months, 3 weeks

1
0
0 0

Theme One Program • Motivation

by Jon Awbrey

Theme One Program • Motivation 1 • https://inquiryintoinquiry.com/2024/06/03/theme-one-program-motivation-1-b/ All, The main idea behind the Theme One program is the efficient use of graph‑theoretic data structures for the tasks of “learning” and “reasoning”. I am thinking of “learning” in the sense of learning about an environment, in essence, gaining information about the nature of an environment and being able to apply the information acquired to a specific purpose. Under the heading of “reasoning” I am simply lumping together all the ordinary sorts of practical activities which would probably occur to most people under that name. There is a natural relation between the tasks. Learning the character of an environment leads to the recognition of laws which govern the environment and making full use of that recognition requires the ability to reason logically about those laws in abstract terms. Resources — Theme One Program • Overview • https://oeis.org/wiki/Theme_One_Program_%E2%80%A2_Overview Theme One Program • Exposition • https://oeis.org/wiki/Theme_One_Program_%E2%80%A2_Exposition Theme One Program • User Guide • https://www.academia.edu/5211369/Theme_One_Program_User_Guide Survey of Theme One Program • https://inquiryintoinquiry.com/2024/02/26/survey-of-theme-one-program-6/ Regards, Jon cc: https://www.academia.edu/community/LY0BKD cc: https://mathstodon.xyz/@Inquiry/112554698776574635

9 months, 4 weeks

1
5
0 0

Re: [ontolog-forum] Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomynote

by John F Sowa

Alex, I like your note below, which is consistent with my previous note that criticized your earlier note, If this is your final position, I'm glad that we agree. As for translations from one language to another, we can't even depend on humans. When absolute precision is essential, it's important to produce an "echo" -- a translation from (1) the user's original language (natural or formal) to (2) the computer system's internal representation to (3) the same language as the user's original, and (4) A question "Is this what you mean?" John ---------------------------------------- From: "Alex Shkotin" <alex.shkotin(a)gmail.com> John and All, I began but not finished yet one report [1] of LLM ability to verbalize formal language, in this case OWL2. The bad places have yellow and red colors. And here [2] is an example of our dialog. But the summary for me is clear: we can't trust LLM even for "translations from one language (natural or artificial) to another" It is mostly correct but sometimes unexpectedly wrong.⚽ Id est even in this case we need "Revision" before to give LLM output to decision making. Alex

10 months

1
0
0 0

Re: [ontolog-forum] Trends in Integration of Knowledge and Large Language Models: A Survey and Taxonomy

by John F Sowa

Alex, No, that third method is NOT what I was saying. ALTHOUGH their third method (below) may use precise methods, which could include ontology and databases as input, their FINAL process uses LLM-based methods to combine the information. (See Figure 4 below, which I copied from their publication.) When absolute precision is required, the final reasoning process MUST be absolutely precise. That means precise methods of logic, mathematics, and computer science must be the final step. Probabilistic methods CANNOT guarantee precision. Our Permion.ai company does use LLM-based methods for many purposes. But when absolute precision is necessary, we use mathematics and mathematical logic (i.e. FOL, Common Logic, and metalanguage extensions). Wolfram also uses LLMs for communication with humans in English, but ALL computation is done by mathematical methods, which include mathematical (formal) logic. Kingsley has also added LLM methods for communication in English. But his system uses precise methods of logic and computer science for precise computation when precision is essential. For examples of precise reasoning by our old VivoMind company (prior to 2010), see https://jfsowa.com/talks/cogmem.pdf . Please look at the examples in the final section of those slides. The results computed by those systems (from 2000 to 2010) were far more precise and reliable than anything computed by LLMs today. I am not denying that systems based on LLMs may produce reliable results. But to do so, they must use formal methods of mathematics, logic, and computer science at the final stage of reasoning, evaluation, and testing. John ---------------------------------------- From: "Alex Shkotin" <alex.shkotin(a)gmail.com> Sent: 6/7/24 3:53 AM John, Please! And shortly. If I want a very reliable LLM, I have to train it myself. JFS: "That article is interesting. But without an independent method of testing and verification, Figure 4 is DIAMETRICALLY OPPOSED to the methods we have been discussing and recommending in Ontolog Forum for the past 20 or more years." But this green box is all about your point. One interesting point from one talk on the Internet is that Huge Language Models (from ChatGPT to now) use ALL World Wide Available Knowledge we have and it is not enough to make it good. But we do not have more for them🙂 Alex чт, 6 июн. 2024 г. в 22:13, John F Sowa <sowa(a)bestweb.net>: Alex, Thanks for the reference to that article. But the trends it discusses (from Dec 2023) are based on the assumption that all reasoning is performed by LLM-based methods. It assumes that any additional knowledge is somehow integrated with or added to data stored in LLMs. Figure 4 from that article illustrates the methods the authors discuss: Note that the results they produce come from LLMs that have been modified by adding something new. That article is interesting. But without an independent method of testing and verification, Figure 4 is DIAMETRICALLY OPPOSED to the methods we have been discussing and recommending in Ontolog Forum for the past 20 or more years.

10 months

1
0
0 0

2025

2024

2023

2022

2021

CG