Gemma 3 Sets New Bar for Open Weight AI Models
Google DeepMind releases Gemma 3, delivering frontier-class performance in an open weight model that runs on a single GP…
🏷️ Large Language Models
90 articles about 'Large Language Models'
OpenAI o3-mini Scores Gold on Math Olympiad
OpenAI's o3-mini reasoning model achieves gold medal-level performance on International Math Olympiad problems, signalin…
OpenAI Launches GPT-5 Turbo With Multimodal AI
OpenAI unveils GPT-5 Turbo featuring native multimodal reasoning across text, image, audio, and video inputs.
Google DeepMind Launches Gemini 2.5 Ultra
Google DeepMind unveils Gemini 2.5 Ultra, its most powerful AI model yet, featuring a 1-million-token context window and…
Why LLM Hallucination Is Inevitable, Not Fixable
New research and theoretical analysis suggest hallucination is a fundamental, mathematical limitation of large language …
DeepSeek V4 Approaches Frontier-Level AI Performance
DeepSeek's latest V4 model closes the gap with top Western AI systems, raising questions about the future of the AI race…
Talking to Transformers: How We Communicate With AI
The art and science of communicating with transformer-based AI models is reshaping how developers and users interact wit…
DeepSeek V4 Benchmarks Show 8-Month Gap With US AI Leaders
DeepSeek's latest V4 model series arrives 15 months after R1, but benchmark comparisons suggest it still trails top US m…
Beyond Conversation: Seven Unconventional Ways to Use Large Language Models
Large language models are capable of far more than chat-based Q&A. This article outlines seven unconventional LLM applic…
New Research on LLM Reasoning Controllability: The Tug-of-War Between Compliance and Sensibility
A new arXiv paper explores whether fundamental reasoning modes such as deduction, induction, and abduction can be decoup…
The Semantic Structure of LLM Feature Spaces Turns Out to Be Highly Aligned with Human Psychology
A new study reveals that the geometric relationships of semantic features in the hidden states of large language models …
New LLM Benchmarking Framework: Human-AI Collaborative Assessment of Mathematical Competency
A new study proposes a 'human-in-the-loop' benchmarking framework that systematically evaluates the performance of multi…