Sia’s GEN-1 robot, powered by a GPT-3-style large language model, claims a jump in task-success rate from 64% to 99%, signaling a shift from simple perception-execution to cognitive decision-making, while the article scrutinizes the definition of success, cost, safety, and industry impact.
The article traces the development of GPT series—from the 2017 Transformer breakthrough, through GPT‑1, GPT‑2, and GPT‑3’s 175 billion parameters, to later models like Codex and ChatGPT—highlighting key papers, architectural choices, and the surprising role of OpenAI’s decoder‑only approach.
This article investigates whether replacing traditional ID-based item encoders with massive LLMs such as GPT‑3 improves recommendation performance, by conducting extensive experiments on three real‑world datasets, analyzing performance limits, generality of item representations, and comparing against ID‑based and prompt‑based methods.
This article analyzes the Prophet framework, which leverages a traditional VQA model to generate answer candidates and in‑context examples that prompt GPT‑3, achieving state‑of‑the‑art performance on the challenging OK‑VQA and A‑OKVQA benchmarks.
Microsoft researchers fine‑tuned GPT‑3.x models with LoRA on over 40,000 cloud incident records, evaluated them with six NLP metrics and human interviews, and found that LLMs can generate root‑cause analyses and mitigation steps comparable to BERT models, especially for machine‑detected failures.
This article explains what ChatGPT is, how it builds on natural-language-processing and the Transformer-based GPT series, details its model-size growth, architectural enhancements, multilingual support, and walks through the tokenization-to-generation pipeline that enables coherent AI-driven conversations.
The article surveys major large language models—including GPT‑3, T5, LaMDA, Jurassic‑1, MT‑NLG, Gopher, Chinchilla, PaLM, U‑PaLM, OPT, LLaMA, BLOOM, GLM‑130B, and ERNIE 3.0 Titan—explains their architectures, scaling trade‑offs, and then details instruction‑fine‑tuned variants such as T0, FLAN, GPT‑3.5, ChatGPT, GPT‑4, Alpaca and ChatGLM, providing references for further study.
This article provides an overview of ChatGPT’s rapid rise, OpenAI’s founding, the evolution of GPT models up to GPT‑3, the data‑driven training process, model capabilities and limitations, and practical guidance for users, highlighting the interplay between open‑source research and commercial deployment.
This article explains the concept of in‑context learning in large language models, presents experimental evaluations such as copy‑output, date‑formatting, and label‑remapping tasks, and discusses a recent theoretical analysis that links attention layers to implicit gradient‑based fine‑tuning, highlighting why model scale and data volume matter.
In an InfoQ interview, JD Technology senior algorithm scientist Wu Youzheng reviews the rapid advances of natural language processing in 2020—including GPT‑3, multimodal dialogue, knowledge‑enhanced pre‑training, and knowledge graphs—while outlining the most promising research directions and practical challenges for the coming year.
This article demonstrates how GPT‑3 can translate plain‑English questions into SQL statements, starting with simple counts and progressing to multi‑table joins, while highlighting prompting techniques, limitations, and the impact of model variants such as GPT‑3 Instruct.
