Education
Saturday, June 13, 2026
9 posts
@Google's Gemini-SQL2, powered by 3.1 Pro, scores 80.04% execution accuracy on BIRD text-to-SQL leaderboard.
https://www.marktechpost.com/2026/06/12/google-releases-gemini-sql2-gemini-3-1-pro-text-to-sql-scores-80-04-on-bird-single-model-leaderboard/
SpatialClaw rethinks action interfaces for agentic spatial reasoning in 3D environments.
https://arxiv.org/abs/2606.13673v1
Mana introduces a new approach for dexterous manipulation of articulated tools in robotics.
https://arxiv.org/abs/2606.13677v1
@OpenAI lets Codex users bank rate-limit resets and trigger them manually instead of watching them expire.
https://the-decoder.com/openai-kicks-off-the-ai-price-wars-with-flexible-rate-limit-resets-for-its-codex-coding-agent/
New RAG approach trains models to reason by analogy, not just keyword similarity.
https://arxiv.org/abs/2606.13680v1
Compliance Theatre, Act 28.
The Annex III review concluded that the model lacked sufficient GPAI literacy.
https://garymarcus.substack.com/p/breaking-news-us-commerce-department
@AnthropicAI's Claude Fable 5 costs twice as much for 5.7 percent more performance, topping the AI Intelligence Index. https://the-decoder.com/anthropics-claude-fable-5-costs-twice-as-much-for-5-7-percent-more-performance/
US government directive to suspend foreign access to Fable 5 and Mythos 5 under national security authorities.
https://www.lesswrong.com/posts/f5avt6eEzkGJJqcCe/us-government-directive-to-suspend-access-to-fable-5-and
SpaceX's IPO tests trillion-dollar tech valuations, signaling potential for OpenAI and Anthropic. Investor Steve Rattner questions sustainability of such scales. Source: https://www.bloomberg.com/news/videos/2026-06-13/can-tech-justify-a-trillion-dollar-valuation-video