惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

酷 壳 – CoolShell
酷 壳 – CoolShell
aimingoo的专栏
aimingoo的专栏
P
Proofpoint News Feed
宝玉的分享
宝玉的分享
MyScale Blog
MyScale Blog
The GitHub Blog
The GitHub Blog
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
月光博客
月光博客
量子位
博客园 - 司徒正美
V
V2EX
I
InfoQ
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Vercel News
Vercel News
H
Hackread – Cybersecurity News, Data Breaches, AI and More
美团技术团队
N
Netflix TechBlog - Medium
L
LangChain Blog
IT之家
IT之家
Blog — PlanetScale
Blog — PlanetScale
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Stack Overflow Blog
Stack Overflow Blog
A
About on SuperTechFans
Microsoft Azure Blog
Microsoft Azure Blog

Hacker News - Newest: "AI"

AI can't read an investor deck AI as an attorney? Student uses ChatGPT, Gemini to sue UW over alleged racial discrimination Hacking MCP Servers in AI Systems – The Rug Pull: Tool Changes After Approval GitHub - MeepCastana/KubeezCut: Free Web based video editor Can AI judge journalism? A Thiel-backed startup says yes, even if it risks chilling whistleblowers Coming soon: 10 Things That Matter in AI Right Now DARPA built an AI to fact-check enemy weapons claims What explains heterogeneity in AI adoption? When AI Meets Muscle: Context-Aware Electrical Stimulation Promises a New Way to Guide Human Movements - Department of Computer Science AI Changed How We Build. It Did Not Change What Matters. Linux rules on using AI-generated code - Copilot is OK, but humans must take 'full responsibility for the… Meta spins up AI version of Mark Zuckerberg to engage with employees Code Mode: Let Your AI Write Programs, Not Just Call Tools | TanStack Blog GitHub - Delavalom/graft: Go framework for building AI agents. Type-safe tools, multi-provider (OpenAI, Anthropic, Gemini, Bedrock), zero vendor SDKs. India's TCS tops estimates, says new AI models did not dent services demand Gen Z's fading AI hype Strong feeling: we are in a folded AI reality GitHub - machinarii/total-recall-catalog: A reference catalog of latest knowledge retrieval, memory & RAG systems GitHub - mensfeld/code-on-incus: Give each AI agent its own isolated machine with root, Docker, and systemd. Active defense detects and stops threats automatically.. Quantization, LoRA, and the 8% Problem: Benchmarking Local LLMs for Production AI Iran war: We spoke to the man making Lego-style AI videos that experts say are powerful propaganda Powell, Bessent discussed Anthropic's Mythos AI cyber threat with major U.S. banks GitHub - immartian/bellamem: Persistent belief-graph memory for AI agents. Retrieves decisive context by importance — not recency, not RAG, not /compact. recursive-mode: The Repo-Native Operating System for AI Engineering After the attack on Sam Altman's home, will AI CEO's go on the offensive? The biggest advance in AI since the LLM Opus 4.6 vs GPT 5.4 One Prompt Unity World Generation Test “AI polls” are fake polls Client Challenge Can AI be a 'child of God'? Inside Anthropic's meeting with Christian leaders
GitHub - pekkalepola/colibri-clf: COLIBRI — Concept Libra...
PekkaLepola · 2026-05-10 · via Hacker News - Newest: "AI"

Single Concept Format Specification v0.94

This repository publishes the CLF 0.94 pre-standard draft — the file format for a single concept in the COLIBRI Concept Library.

Concept Purity Rule

A concept file expresses only what a concept IS and what is perceptually typical of it. It contains no detection logic, no rules, no algorithms, no thresholds, and no references to other concepts. All interpretation belongs exclusively to the control layer.

A concept file MUST NOT contain any structure, field, reference, or wording that can be interpreted as prescribing, constraining, guiding, or implying any behavior of the control layer, inference layer, or any system that consumes the concept.

Format Status

This specification is published at version 0.94. The format is intentionally not yet final.

The goal of the COLIBRI Concept Library is broad adoption across industries and systems. A format that is to become a shared standard must be shaped by those who will use it.

Status of This Document

This document specifies the CLF 0.94 pre-standard draft — the file format for a single concept in the COLIBRI Concept Library. CLF is one component of a broader standard currently under development, which will also cover registration practices, concept identifier naming, versioning at the library level, and governance.

This is not the final standard. The final specification will be decided together with early adopters, implementers, and domain experts.

No rights to implement, deploy, or integrate this format are granted. All commercial and technical usage requires a separate license from the rights holder.

Underlying invention protected by Finnish Utility Model No. 13913.

Contents

File Description
clf_concept_spec_v0.94.md Single concept format specification v0.94
Concept_library_wp.pdf White paper — full theoretical foundation
What_is_Concept_Library.md Introductory guide — what the Concept Library is and why it exists
clf_control_layer_v1.py Control layer reference implementation — loads CLF concepts, matches queries, produces audit trail
demo.py Runnable demo — five example queries with full semantic audit trail output
DEMO.md Usage guide for the control layer reference implementation
C-003.partial_discharge/ Example concept — partial discharge (electrical insulation phenomenon)
C-005.cat/ Reference concept — domestic cat (Felis catus)

Reference Concept

C-005.cat is a fully realized reference implementation of the CLF 0.94 format, covering all seven perceptual modalities with genuine data:

Modality Content
Linguistic English canonical definition, exclusions, semantic scope
Visual Parametric SVG silhouette (Bezier ears, vertical-slit pupils, whiskers, tail arc)
Acoustic Synthetic FLAC prototype — FM sweep 500→900→450 Hz with 4 harmonics, ADSR envelope
Signal Real MFCC-13, spectral centroid (1392 Hz), rolloff (2738 Hz), computed from prototype
Haptic Fur texture, temperature, body density, motion resistance
Chemical Olfactory profile — 2-acetyl-1-pyrroline, pheromone classes
Electromagnetic Coat static triboelectric, infrared thermal profile

The concept also includes integrity data: a full SHA-256 hash tree over all payload files and a Merkle root as the container hash.

Format History

Version Date Change
0.9 2026-04-30 Initial pre-standard draft
0.92 2026-05-05 Added SVG and WAV to allowed formats; concept_kind moved exclusively to taxonomy.json; definition_language required in manifest; i18n translations removed (belong to control layer); English mandated as canonical language
0.93 2026-05-06 biological/cognitive removed from concept_kind (domains, not kinds); prototype_spectral_features moved to acoustic.json; section numbering fixed; container structure shows prototype file paths
0.94 2026-05-06 constraints.json removed entirely — classification conditions belong to control layer; labels restricted to English only; control layer boundary table extended

Control Layer Reference Implementation

The control layer is the component that decides what a given input IS — and explains why. A concept file describes what something is. The control layer does the matching.

clf_control_layer_v1.py is a reference implementation of the control layer included in this repository. It loads any CLF concept folder, accepts multi-modal queries (text, acoustic features, signal data), and returns the best-matching concept together with a full semantic audit trail.

python demo.py

See DEMO.md for full usage instructions and code examples.

Production use, commercial deployment, and integration into third-party systems requires a separate license from the rights holder.

White Paper

Concept Library by Pekka Lepola, 2026.

The white paper presents the theoretical foundation, architectural consequences, EU AI Act implications, and a proof-of-concept semantic decision trace.


© 2026 Pekka Lepola. All rights reserved.

The concept library and control layer invention is protected by Finnish Utility Model No. 13913.

No part of this specification may be used, reproduced, or implemented without explicit written permission from the rights holder.

Licensing & Pilot Program: Organizations interested in piloting the COLIBRI Concept Library may acquire a paid pilot license. Contact: conceptlibrary.eu@gmail.com