목차
Preface
“
In this essay, I propose a research agenda that I believe will eventually lead to coding AIs that have superhuman-level ability, are hallucination -free, and safe. To achieve our goal will require a lot of further innovation, experimentation, and hard work. But I fully believe this is the most promising direction. (…omitted…)
gasstationmanager.github.io/ai/2024/11/04/a-proposal.html
See also
목차 관련 글 Siren's song in the AI ocean: A survey on hallucination in large language models “Fact-conflicting hallucination, where LLMs generate content that is not faithful to established world knowledge.” 열린 도메인 환각 “Open-domain hallucination is a hallucination where a model confidently generates false information about the wo…” AI Hallucinations: A Misnomer Worth Clarifying “Confabulation: AI generated responses that sound plausible but are, in fact, incorrect.” Survey of hallucination in natural language generation “In this survey, we thus provide a broad overview of the research progress and challenges in the hallucination p…” A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions “Nevertheless, LLMs are prone to hallucination, generating plausible yet nonfactual content.” Why Language Models Hallucinate 관련 절: Post-training and hallucination LLMs will always hallucinate, and we need to live with this “This work introduces the concept of Structural Hallucination as an intrinsic nature of these systems.” A comprehensive taxonomy of hallucinations in Large Language Models “Furthermore, the report examines cognitive and human factors influencing hallucination perception, surveys eval…” Language Models Know More Than They Show: Exploring Hallucinations From the Model's Viewpoint “Large language models (LLMs) often produce errors, including factual inaccuracies, biases, and reasoning failur…” GPT-4 technical report 관련 절: Memo Hallucinating faces “The additional pixels are, in effect, hallucinated.” 환각 (AI) “In psychology, a hallucination is a false sensory perception in the absence of an external stimulus, as distinc…” Semi-automated coding for qualitative research “Semi-Automated Coding for Qualitative research: A User-Centered Inquiry and Initial Prototypes” Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AI “This review presents a comprehensive analysis of two emerging paradigms in AI-assisted software development: vi…” HtmlRAG: HTML is Better Than Plain Text for Modeling Retrieved Knowledge in RAG Systems “Retrieval-Augmented Generation (RAG) has been shown to improve knowledge capabilities and alleviate the halluci…” Training LLMs for Honesty via Confessions “To demonstrate the viability of our approach, we train GPT-5-Thinking to produce confessions, and we evaluate i…” 린 (프로그래밍 언어) “Lean is a functional programming language that makes it easy to write correct and maintainable code.” 바이브 코딩 “2025-03-19 - Not all AI-assisted programming is vibe coding (but vibe coding rocks)” AI and science “2026-02-05 - GPT-5 lowers the cost of cell-free protein synthesis” Self determination 관련 절: [[Lean software development]]