8x Terminal Performance Gains: NVIDIA's Data Recipe Lets 32B Beat 480B
AI Agents
18 posts in AI Agents
Mirage of Synthesis: DREAM's Agentic Framework Catches What Static Benchmarks Miss
80.3 on ScreenSpotPro: GUI-Owl-1.5 Sets New Bar for Open-Source GUI Agents
When Should AI Agents Ask for Help? CMU's CowCorpus Maps Four Human Collaboration Styles
Even GPT-5 Fails at Discovery: OdysseyArena Exposes the Inductive Bottleneck in LLM Agents
Prompt Fatigue Solved: Vibe AIGC Turns Users Into 'Commanders' of Multi-Agent Creative Workflows
Google Introduces Agentic Vision: Gemini 3 Flash Now Zooms, Annotates, and Investigates Images
260% Better at Catching Moving Objects: DynamicVLA Solves Robot Latency Problem
15-Hour Agent Runtimes Solved: Idea2Story Precomputes Research Knowledge Offline
73% on BrowseComp: Meituan's 560B Open-Source Model Leads Agentic Benchmarks
56.7% on OSWorld: EvoCUA's Evolutionary Training Beats Closed-Source Computer Use Agents
SimpleMem gives LLM agents 30x cheaper memory with 26% better recall
Microsoft's Agent Lightning Decouples RL Training from Agent Logic, Enabling Fine-Tuning of Any AI Agent with Zero Code Changes
Agent Memory Fragmentation Solved: EverMemOS Achieves 93% on LoCoMo via Engram-Inspired Lifecycle
Agent Memory Loss Solved: InfiAgent's File-Centric Architecture Enables Unlimited Runtime
DeepResearchEval: Benchmark Shows Gemini Leads Quality, Manus Wins Factual Accuracy
Zero Training Data, Full Performance: Dr. Zero Matches Supervised Search Agents
Gold Medal at IMO and IOI: DeepSeek-V3.2 Matches GPT-5 with Open Weights
