pamperherselfHow LLMs Learn in Unseen Contexts: In-Context Learning and Induction HeadsWhile researching the OECD Global AI Policy,Rankings and Germany AI Development Report , I came across the “Reasonable AI” (RAI) research…12h ago12h ago
pamperherselfOECD Global AI Policy,Rankings and Germany AI Development ReportThis article is based on my interpretation of the OECD reports: OECD National AI Policy 2021 and AI Review of Germany 2023. These reports…1d ago1d ago
pamperherselfOpenAI Official Guide: 3-Step Process to Reduce Hallucinations and Increase LLM Answer AccuracyI’ve always been concerned about the accuracy of large language model (LLM) responses. Recently, I was considering canceling my ChatGPT…Sep 19Sep 19
pamperherselfLife Is Sometimes Like Riding a Bicycle — To Keep Balance, You Must Keep MovingEvery day after my midday nap, I ride a shared bike to work. After each ride, a system screen pops up asking if the brakes worked properly…Sep 18Sep 18
pamperherselfLauren Manoogian: Casual Elegance with Peruvian Craftsmanship, But Alpaca Wool is Just ThatAs autumn sets in, it’s time for cashmere season again. Starting with Lauren Manoogian, let’s dive into premium designer brands that focus…Sep 15Sep 15
pamperherselfGPT4+prompt ≈ GPT o1? It’s 2024, and We’re Still Fine-Tuning and Self-Training Models?This morning, as I started work and opened ChatGPT, I habitually switched from GPT-4 mini to GPT-4, only to notice some unfamiliar models —…Sep 13Sep 13
pamperherselfAvoiding the Noise of Fragmented Information: No Short Videos, No Brief PostsLast week, I tried updating on Red and other fragmented platforms, thinking that posting the same content on multiple platforms would…Sep 13Sep 13
pamperherselfThe Fabric Enthusiast Spent 4 Days Exploring Over 300 Brands to Curate the 30 Best Niche Cashmere…In Beijing, if you’re looking to buy brands that exude Quiet Luxuryyour options are limited to SKP and Lane Crawford, unlike Shanghai where…Sep 13Sep 13
pamperherselfImproving RAG Retrieval Quality: 4 Optimization Guidelines from ShortwavePreviously, I wrote an article titled Hands-on Experience with AnythingLLM: RAG Tools Seem Superficial and Too Rigid, which touched on…Sep 6Sep 6
pamperherselfEvaluating LLMs : 5 Strategies to Make GPT More Compliant (title generator Prompt Tips Included)I have never delved deeply into evaluating large models before. In the past, I would often skip evaluation-related topics. However, when I…Sep 6Sep 6