Tag: research
All the articles with the tag "research".
-
This New LLM Agent Framework Hits SOTA With Just $18 of Training (Yes, Really)
• 1 min readAn open-weight agent framework just beat closed models on tough web + reasoning benchmarks using pocket-change training costs—and that shoul
Read more -
This Sneaky 'Data Poison' Trick Could Make Stealing Your AI Model Totally Useless
• 1 min readResearchers built a tool that poisons stolen AI data graphs, tanking thief models to 5% accuracy while yours stays perfect.
Read more -
Generative AI Growth Isn’t Just Hype — Researchers Want Reliability and Explainability Now
• 1 min readGenerative AI is booming at ~47% CAGR — but the real conversation has shifted to reliability, explainability, and job impacts.
Read more -
LLMs Are Teaching Materials Science to Be Autonomous — and That’s Actually Useful
• 1 min readLarge language models are moving from chat to the lab bench, automating materials discovery with surprising explainability.
Read more -
Google Gemini 3 Deep Research Agent Released
• 1 min readGoogle's Gemini 3 Deep Research Agent hits 46.4% on Humanity's Last Exam, autonomously handling multi-step tasks—empowering devs with Intera
Read more -
OpenAI GPT-5.2 Launch: Frontier LLM Breakthrough
• 1 min readOpenAI's GPT-5.2 frontier model rushes to rival Gemini 3 with 3T parameters, 256k context, and superior reasoning—redefining LLM capabilitie
Read more -
A-HMAD: AI Agents Debate for Superior Math Reasoning
• 1 min readA-HMAD framework lets AI agents debate to boost LLM mathematical reasoning, outperforming single models on GSM8K and chess—revolutionize com
Read more -
Agent Lightning: RL for AI Agents Without Code Changes
• 1 min readMicrosoft's Agent Lightning adds reinforcement learning to LLM agents with zero code rewrites, enabling self-improving AI for complex tasks—
Read more -
NeurIPS 2025 Best Paper: Gated Attention in LLMs
• 1 min readNeurIPS 2025 Best Paper reveals gated attention breakthrough for LLMs, boosting stability and long-context performance on 15B models—essenti
Read more -
Agent Lightning Adds RL to AI Agents Seamlessly
• 1 min readMicrosoft's Agent Lightning integrates reinforcement learning into LLM agents with zero code changes, enabling self-improving AI for complex
Read more -
OpenAI Circuit-Sparsity Optimizes LLMs by Deactivating Neural Circuits
• 1 min readNew OpenAI technique cuts LLM inference FLOPs 50-70% by pruning low-impact circuits, enabling faster, cheaper deployment with little accurac
Read more -
OpenAI Launches GPT-5.2 Series with Enhanced Reasoning and Coding
• 1 min readOpenAI released GPT-5.2 Pro and Thinking models, boosting coding, enterprise tasks, long-context understanding amid competition with Google.
Read more
