Implicit Knowledge Is a Liability

AI 코더는 프로젝트에 대한 암묵적 지식이 부족해 회귀 버그 발생 위험이 크다. 암묵적 지식을 명시적이고 재현 가능한 테스트로 전환하는 것이 AI 시대에 신뢰성 높은 소프트웨어 개발의 핵심이다. 문서나 주석 같은 명시적 지식도 종종 오래되어 AI가 무시할 수 있으므로, 행동 검증 중심의 테스트가 필수적이다. AI가 테스트 작성에 도움을 줄 수 있으나, 엄격한 감독이 필요하다. AAA, Given-When-Then 같은 테스트 패턴과 공용 API 중심 테스트가 여전히 최선의 방법이다.

https://news.ycombinator.com/item?id=48108956

#aicoding #softwaretesting #regression #implicitknowledge #testautomation

Implicit Knowledge Is a Liability | Hacker News

I feel like I’m waiting for something that isn’t going to happen.

https://cezar.vivaldi.net/2024/11/11/implicit-knowledge/

Cezar Danilevici • Implicit Knowledge

I feel like I’m waiting for something that isn’t going to happen.

Cezar Danilevici
Worst Japanese Learning Mistake You Might Be Making Right Now

YouTube

... cont'd

* KG often equipped with ontological schema (including rules, constraints, ontologies) ensuring quality, enabling easier knowledge access, supporting reasoning, etc.
* In this part, we introduce topics that LLMs are applied to learn ontological schemas & to manage ontologies
* OntoGPT: https://monarch-initiative.github.io/ontogpt

#OntoGPT #OntoChatGPT #LLM #LargeLanguageModels #GPT #ChatGPT #GPT3 #GPT4 #KnowledgeGraphs #epistimology #ImplicitKnowledge #ExplicitKnowledge #ParametricKnowledge #ontology

OntoGPT

[megathread] LLM & knowledge Graphs

Large Language Models & Knowledge Graphs: Opportunities, Challenges
https://arxiv.org/abs/2308.06374

* position paper: https://en.wikipedia.org/wiki/Position_paper
* hybrid representation of explicit knowledge (knowledge graphs) & parametric knowledge (LLM)

#LLM #LargeLanguageModels #LanguageModels #NLP #NaturalLanguageProcessing #GPT #ChatGPT #GPT3 #GPT4 #KG #KnowledgeGraph #KnowledgeGraphs #epistimology #knowledge #ImplicitKnowledge #ExplicitKnowledge #ParametricKnowledge #ontology

Large Language Models and Knowledge Graphs: Opportunities and Challenges

Large Language Models (LLMs) have taken Knowledge Representation -- and the world -- by storm. This inflection point marks a shift from explicit knowledge representation to a renewed focus on the hybrid representation of both explicit knowledge and parametric knowledge. In this position paper, we will discuss some of the common debate points within the community on LLMs (parametric knowledge) and Knowledge Graphs (explicit knowledge) and speculate on opportunities and visions that the renewed focus brings, as well as related research topics and challenges.

arXiv.org