Lmst

This month's Acuitas diary features expanded work on conjunction group parsing, including a bit of nitty-gritty on text parser data structures and design choices.

https://writerofminds.blogspot.com/2026/01/acuitas-diary-92-january-2026.html

#ArtificialIntelligence #AI #chatbots #symbolicAI #NLP

Clip art of a classic blank scroll, curled in opposite directions at both ends.

Et si le vocabulaire que tu utilises tous les jours n’était pas neutre ?
Quand tu dis « appeler une fonction », « appeler une API », « appeler un service », qu’est-ce que tu acceptes implicitement comme architecture ?
Un monde de processus qui tournent en permanence, de sockets ouverts, de files d’attente, de watchers qui surveillent, de latences à gérer ?
Et si, rien qu’en changeant un mot, tu pouvais sortir de cette prison invisible ?
#SymbolicAI
#Neurosymbolic
#AIArchitecture

Que se passerait-il si, au lieu d’appeler, tu commençais à invoquer ?

Last Acuitas diary blog of the year! Featuring a miscellany of little quirks I worked out, refactoring, and unfinished business. https://writerofminds.blogspot.com/2025/12/acuitas-diary-91-december-2025.html
#ArtificialIntelligence #AI #SymbolicAI #NLP #chatbots

Silhouette of a male or generic human face, looking toward the viewer, with one hand cradling the chin as if in deep thought.

This month, I worked out how to tell Acuitas the simplest detective story I could think of. https://writerofminds.blogspot.com/2025/11/acuitas-diary-90-november-2025.html
#AI #ArtificialIntelligence #chatbots #NLP #SymbolicAI

Frederic Dorr Steele's illustration for the Sherlock Holmes story "The Adventure of the Dying Detective" as published in The Strand Magazine. It's a black-and-white drawing of what looks to be a study; there's a table in the center, with a number of books and bottles on top, and a fireplace in the background. The lighting is dim, and a lamp on the table throws heavy contrast on three people standing around it. They all look rather serious or intense. One man is bending over the table and leaning on it with one hand; in his other hand he holds one of the books. He is facing the remaining two men who are at the opposite corner of the table.

Could #SymbolicAI Unlock Human-like Intelligence?

Combining newer neural networks with older #AI systems could be the secret to building an AI to match or surpass human intelligence

https://www.nature.com/articles/d41586-025-03856-1

#Innovation

In Acuitas news, I dust off episodic memory and try to unify it with the more recently developed narrative architectures. These open new possibilities for methods of summarizing, characterizing significance, and selectively forgetting memories.
https://writerofminds.blogspot.com/2025/10/acuitas-diary-89-october-2025.html
#AI #ArtificialIntelligence #SymbolicAI #NLP #chatbots

A clutter of assorted books stacked on a concrete floor. "Disorganized books piling one another", by Ibrahim Husain Meraj via Wikimedia Commons. See source for license.

Tensor Logic: Ngôn ngữ mới kết hợp AI neural & symbolic đơn giản hóa bằng tensor equation. Giải quyết hạn chế PyTorch/LISP. Ewent transformers, logic, thèm học. 📚TAG: #AI #TensorLogic #DeepLearning #SymbolicAI #ResponsibleTech #HackedNews

https://www.reddit.com/r/LocalLLaMA/comments/1o891o4/tensor_logic_the_language_of_ai/

Acuitas' Text Parser can finally (in theory) handle all the grammatical structures in the test sets! https://writerofminds.blogspot.com/2025/09/acuitas-diary-88-september-2025.html
#ArtificialIntelligence #AI #chatbots #NLP #NLU #symbolicAI

Three pie charts showing the percentage correct and incorrect for the three test sets: "Magic Schoolbus: Inside the Earth" (53%/47%), "Out of the Dark" (54%/46%), and "Log Hotel" (81%/19%).

LLMs können viel – aber beim symbolischen Planen scheitern sie oft.
Ein neues Paper von MIT CSAIL & Microsoft zeigt, wie Instruction Tuning mit logischen Chain-of-Thought-Schritten und externem Validator (VAL) LLMs zu verlässlichen Planern macht.

Die Ergebnisse: bis zu 94 % gültige Pläne statt 28–35 % in den Baselines.
https://arxiv.org/abs/2509.13351
#AI #LLM #Planning #AgenticAI #SymbolicAI #HybridAI #TrustworthyAI

A flowchart illustrating the PDDL-INSTRUCT approach, consisting of two main training phases: Initial Fine-tuning and Chain-of-Thought (CoT) Instruction Tuning, along with an evaluation phase. It shows the processes involving datasets, a pre

@jannem @cmccullough

"When technology works noone calls it AI anymore."

There's also another subfield in AI called Automated Reasoning and Automated Theorem Proving, used to proof the correctness of software:
https://en.wikipedia.org/wiki/Automated_reasoning

...or Model-based reasoning (the hype term "AI model" has made this term unsearchable, unfortunately)

They all fall under the umbrella term Symbolic #AI - this is 100% reliable AI, but noone is talking about it!

https://floss.social/@janriemer/114454349034565839

#SymbolicAI

Only a few days left to apply!

If you are interested in logic, decision-making, reasoning under uncertainty and statistics, apply by the end of this month for an opportunity to work with me, dr. Sicco Verwer and dr. Fabian Mies at Delft University of Technology!

Application deadline: 31 August 2025

https://careers.tudelft.nl/job/Delft-PhD-Position-Symbolic-AI-and-Reasoning-Under-Uncertainty-2628-CD/824585702/

#AcademicJobs
#AcademicMastodon
#GetFediHired
#AcademicJob
#SymbolicAI
#Statistics
#AI
#ConstraintProgramming
#CombinatorialOptimisation
#SensitivityAnalysis
#FormalMethods
#CombinatorialOptimization
#Delft
#TUDelft
#AcademicChatter

One month left to apply!

If you are looking for a PhD position and are interested in working on probabilistic inference, sensitivity analysis, and decision-making, this might be the job for you! We are looking for candidates with a strong background in Computer Science, and ideally also in Mathematics.

Please apply by 31 August. We're looking forward to reading your application!

https://careers.tudelft.nl/job/Delft-PhD-Position-Symbolic-AI-and-Reasoning-Under-Uncertainty-2628-CD/824585702/

#AcademicJobs
#AcademicMastodon
#GetFediHired
#AcademicJob
#SymbolicAI
#Statistics
#AI
#ConstraintProgramming
#CombinatorialOptimisation
#SensitivityAnalysis
#FormalMethods
#CombinatorialOptimization
#Delft
#TUDelft
#AcademicChatter

Hi everyone,

I feel like a re-introduction is long overdue!

My name is Anna, and I'm an assistant professor of Algorithmics at the Delft University of Technology, specialising in combinatorial optimisation, symbolic AI, constraint programming, propositional model counting, operations research and reasoning under uncertainty.

I'm a nerd, a feminist and a traveller, not always in that order.

In my spare time I like to hike and go geocaching. I try to go swing dancing a few times a week. I am a Trekkie. I want to learn how to draw. I am an Indomie and Obsidian enthusiast. Based in the Netherlands, I miss Belgium, Canada and Singapore.

Since a job in academia somehow always is personal, I have chosen to mix professional interactions with the more personal ones on this platform. At least for now. Obviously, my opinions do not necessarily reflect those of my employer yadiyadiyada.

Hope to keep interacting with you all!

#Introduction #AcademicMastodon #Algorithmics #SymbolicAI #CombinatorialOptimisation #ConstraintProgramming #ModelCounting #OperationsResearch #ProbabilisticInference #Geocaching #LindyHop #Jazz #SwingDancing #Hiking #Obsidian #StarTrek #Travel #TUDelft #MastoMiGoreng #Indomie #GNUTerryPratchett #Catstodon #Mastocats #Caturday #ExpatLife #MakanApaToda

Gary Marcus is onto something in here. Maybe true AGI is not so impossible to reach after all. Just probably not in the near future but likely within 20 years.

"For all the efforts that OpenAI and other leaders of deep learning, such as Geoffrey Hinton and Yann LeCun, have put into running neurosymbolic AI, and me personally, down over the last decade, the cutting edge is finally, if quietly and without public acknowledgement, tilting towards neurosymbolic AI.

This essay explains what neurosymbolic AI is, why you should believe it, how deep learning advocates long fought against it, and how in 2025, OpenAI and xAI have accidentally vindicated it.

And it is about why, in 2025, neurosymbolic AI has emerged as the team to beat.

It is also an essay about sociology.

The essential premise of neurosymbolic AI is this: the two most common approaches to AI, neural networks and classical symbolic AI, have complementary strengths and weaknesses. Neural networks are good at learning but weak at generalization; symbolic systems are good at generalization, but not at learning."

https://garymarcus.substack.com/p/how-o3-and-grok-4-accidentally-vindicated

#AI #NeuralNetworks #DeepLearning #SymbolicAI #NeuroSymbolicAI #AGI