Tagged: language-models

22 articles

Latest

GPT-Neo needs just 3 examples to match GPT-3's few-shot trick

GPT-2 flags female pronouns as 85% toxic in new Hugging Face bias test

ChatGPT's Game Design Advice Sounds Great—Until It's Completely Wrong

ChatGPT's Real Secret? It's the Training Stack, Not Just the Model

ChatGPT Writes a Stardew Valley Clone: Why AI Storytelling Still Needs You

Tiny open-source OCR models crush giants at cleaning up AI training data

GPT-2 Is Too Dangerous to Release, Says OpenAI

AI Detectors Fail Up to 29% of the Time When ChatGPT Mimics Your Style

GPT-2 Learns to Cheat on Summaries, Copying Text to Please Human Judges

GPT-2's Final Release Is a 1.5B-Parameter Test Case in AI Safety Theater

DeepMind finds giant models beat tiny data in compute-optimal AI training

GPT-3: 175B params, zero fine-tuning, and humans can't tell

OpenAI Launches GPT-3 API, Inviting Developers Into a Private Beta

OpenAI's RLHF Summarization Model Beats 10x Larger Systems

GPT‑f sneaks new proofs into Metamath's sacred library

OpenAI gives Microsoft exclusive GPT-3 license, API stays open

OpenAI's Big Lesson: API Misuse Looks Nothing Like We Feared

Fill-in-the-middle training doesn't hurt, it helps: new defaults for LLMs

OpenAI and Georgetown Warn Language Models Will Supercharge Propaganda

OpenAI plants flag in Tokyo with 3x faster Japanese GPT‑4

Falcon-H1-Arabic Hits 256K Context, Outperforms Bigger Models with Hybrid Design

Falcon-Arabic 7B beats models 4x its size on Arabic benchmarks