OpenAI × Apollo: How Deliberative Alignment reduces covert overreach
A recent paper systematically proposes an anti-scheming evaluation framework: using a triple stress test involving far-domain tasks, contextual awaren...
Admin •
74
CodeRabbit integrates MCP: AI code review sees the big picture for the first time
CodeRabbit announced the launch of MCP integration, enabling AI code review to aggregate context from GitHub, issues, documentation, and operations to...
Admin •
26
Meta Ray-Ban Display officially released: full-color high-resolution display + EMG Neural Band, AI smart glasses enter a new stage of "wearable and viewable"
Meta has released the new Ray-Ban Display AI smart glasses, which hide content when not needed and display AI-powered information, maps, photos, and m...
Admin •
73
From VR to glasses: Meta Horizon Engine brings "world generation" to portable scenes
Meta announced the Meta Horizon Engine, bringing faster loading times and greater concurrency to the Metaverse and VR scenes, and announced future exp...
Admin •
36
ICPC 2025 Same-Topic Evaluation: OpenAI's Full Score, DeepMind's Gold Medal: What Does This Mean?
OpenAI's reasoning system achieved a perfect 12/12 on the same problem at the 2025 ICPC World Finals, placing first according to official rules. DeepM...
Admin •
112
Qwen3-ASR-Toolkit Released: Breaking the Three-Minute Limit of Qwen3-ASR-Flash, Boosting Hour-Level Audio and Video Transcription
Qwen3-ASR-Toolkit is an open-source CLI tool for Qwen3-ASR-Flash, overcoming the three-minute audio duration limit and enabling efficient transcriptio...
Admin •
159