Prolog et al.

News, info, links, gossip, anything about Prolog, Constraint/(Inductive)Logic programming, Answer Set Programming, and beyond. A @llaisdy project.

[Prolog et al. @PrologInfo]

First draft of a Prolog languages list. Sections on Lambda Prolog, Datalog, and other logic programming languages.

Feedback, PRs, etc., most welcome!

https://t.co/2ooiIqEmAk

[Artificial Intelligence Papers @SciFi]

SCOPE: Compiling Natural-Language Policies into Neurosymbolic Predicate Programs

Arnon Mazza, Elad Levi
https://t.co/TTRAINMSUj [𝚌𝚜.𝙰𝙸 𝚌𝚜.𝙲𝙻]

[Ilya Sergey @ilyasergey]

Huge congratulations to my PhD student Ziyi Yang, who just defended his thesis on Answer Set Programming for Synthesis, Inference, and Search! Very proud of all his achievements (and grateful for him teaching me about ASP along the way). Well done, Dr. Yang!

[𝙶𝚒𝚊𝚗𝚌𝚊𝚛𝚕𝚘 @gfrison]

@HowToPrompt__ @MarkusSchacher I was similarly insulted for my thesis. I created a logic language to lower accidental complexity for also combinatorial problems. You might think “why not just ASP?” https://t.co/BC1pkHm2xM

[takeru.fukushima. @takerufukushima]

@nicobaogim sann

I talked Dr. Hashida Koichi who invented PLR that technology of PDS+encryption.

His todays Interest moves Personal AI(PAI). The paper is here.
https://t.co/3ckvLdn1oh

PAI can reply to human based on the data on PDS and recommend service and products in safely

[Dmitry Vostokov 🇮🇪 @DumpAnalysis]

The initial version of the Prolog WinDbg extension is out there: https://t.co/4xMLzwpT1y The accompanying book will be published later today. Finally, I resumed work that I started more than 15 years ago... https://t.co/EYJktc2hxN

[José A. Alonso @Jose_A_Alonso]

Weekly reads: Oct 19 – Oct 25, 2026.

🇬🇧 English: https://t.co/62SVmokVt2

🇪🇸 Spanish: https://t.co/WxhlondoIM

#AI #AI4Math #ITP #Lean4Challenge #LeanProver #Logic #LogicProgramming #Math #Prolog #RetoLean4

[Dmitry Vostokov 🇮🇪 @DumpAnalysis]

Prolog for Pattern-Oriented Diagnostics: Logic Programming for Memory Dump, Trace, and Log Analysis - a book I asked to be written for me, both to brush up on Prolog and to experiment with my pattern languages. I'm planning to study the paperback version with a pen for possible corrections and new ideas. Full PDF: https://t.co/EwFaIm6hDj

[Mikael Brockman @meekaale]

study the source code to the compiler here, it’s a few thousand lines of Prolog https://t.co/CfZ79OPNO1

[Mikael Brockman @meekaale]

play around with my Fil-C compiler, learn how its capability tracking works under the hood!

it’s an independent full stack compiler that parses C source and emits ABI compatible Fil-C x86 assembly

works enough to build Lua and zlib

https://t.co/h7CyL79pC2

[Prolog et al. @PrologInfo]

Jalog: An interpreter for a Prolog-like language with Java-like arithmetic written in Java

https://t.co/IR7T0gIOe3

[Ari Okkonen @AriOkkonen]

A glimpse to history of AI:
Operation of old "artificial intelligenice" language Prolog can be observed here: https://t.co/qgQmCQEPif

[Ari Okkonen @AriOkkonen]

A glimpse to history of AI:
Operation of old "artificial intelligenice" language Prolog can be observed here: https://t.co/qgQmCQEPif

[Prolog et al. @PrologInfo]

@kibi_dev 😊

[Kibi @kibi_dev]

Kibi's numeric checks are Prolog: https://t.co/BB2OgA8lTD folds each comparison on a variable into an interval and asks if it's empty.

Now those folds are proved exact in Lean 4, and CI runs 200k cases through the Prolog code, the Lean model and a CLP(Q)/CLP(FD) oracle.

[Niraj Munot @niraj_munot]

With @VolrixAI, you can actually talk to your research and backtests.

We apply neurosymbolic principles by combining AI reasoning with code execution, giving you the flexibility to explore ideas, test hypotheses and ask deeper questions with data.

[Extrapolator AI @extrapolatorai]

📰 SCOPE Compiles Agent Policies into Transferable Neurosymbolic Programs

SCOPE compiles natural-language policies into small predicates and code, matching frontier accuracy at a fraction of...

Read more → https://t.co/LEuQ5VFmnq

#AI #AIAgents #AiSafety #InferenceOptimization

[Deepak Kakran @deepakkakran]

The 40-year AI debate (Neural Nets vs. Symbolic Logic) reaches its reckoning in biology.Pure neural AI hallucinates impossible chemistry.Pure symbolic AI shatters under assay noise.The future is Neurosymbolic: Neural Perception + Symbolic Invariant Solvers. 🧵👇

[Peb Ruswono Aryan @pebaryan]

another result by raymotion (raylib)

answer set programming, have you heard about it?

GLM 5.3 Flash (low) from Zcode made it

i kinda like the BGM for a synth

[Kristopher Micinski -- REBORN @krismicinski]

If I want to learn about autoformalization of things in math, what should I read? Is it literally just as simple as it sounds, where you just ask a frontier LLM + agent to formalize a math paper via a relatively simple prompt? Is there anything anyone has written down about it?

[Prolog et al. @PrologInfo]

"MemDerive, a diagnostic benchmark for agent memory over evolving data. It specifies each scenario symbolically, derives exact gold answers with Datalog, ..."

[恒星sun @vintcessun]

Agent memory 最危险的故障，可能不是忘了，而是记着旧状态，还没把变化传下去。

https://t.co/htgpZ2HisS

MemDerive 用 Datalog 生成可追溯的推导链，再把事件历史改写成自然语言，测试 agent 能否从多条记录推出状态，并在新事件到来后及时更新。结果里，常见错误是中间状态虽已存下，却没被检索来回答问题；隐式变化也常未传播。对真实系统而言，记忆需要的不只是向量检索，还要能追踪依赖、管理状态版本，并增量更新推导结果。

[恒星sun @vintcessun]

Agent memory 最危险的故障，可能不是忘了，而是记着旧状态，还没把变化传下去。

https://t.co/htgpZ2HisS

MemDerive 用 Datalog 生成可追溯的推导链，再把事件历史改写成自然语言，测试 agent 能否从多条记录推出状态，并在新事件到来后及时更新。结果里，常见错误是中间状态虽已存下，却没被检索来回答问题；隐式变化也常未传播。对真实系统而言，记忆需要的不只是向量检索，还要能追踪依赖、管理状态版本，并增量更新推导结果。