Congrats to the @MistralAI team on launching Mistral Medium 3.5! This new single 128B dense text-vision model merges instruction-following, reasoning, and coding. Plus, it scored 77.6% on SWE-Bench Verified. The best models run on NVIDIA. Try it out today on
RESEARCH
-
AI Math Capabilities: Conservative Assessment at L2+ Level
By
–
Maybe we should calibrate together. We (generally @tonylfeng
) are more on the conservative side that results in AI for Maths are mostly at L2+ (for now). We haven't seen any L3 result yet. -
SWE-chat: Real-World Coding Agent Interactions Dataset
By
–
"SWE-chat: Coding Agent Interactions From Real Users in the Wild" This paper proposes a real-world coding-agent dataset. It tracks 6K sessions from actual developers, with prompts, tool calls, and line-level human vs. agent authorship. Main result is that agent autonomy is
-
AI-Written Papers at Major ML Conferences in Coming Years
By
–
I wonder how many papers a big AI conferences like @NeurIPSConf
, @ICLR_conf
, and @ICMLconf will be largely AI written in a few years time… -
Harvard Survey: AI Adoption High but Value Lags Without Workflow Integration
By
–
New Survey from Harvard Business Review Analytic Services Finds AI Adoption Remains High, Yet Value May Lag Without Modernization and Workflow Integration https://
appian.com/about/explore/
press-releases/2026/new-survey-from-harvard-business-review-analytic-services-finds-ai-adoption-remains-high
… #AppianWorld -
Eka: A Different Approach to Robotics AI Innovation
By
–
A lot of companies are currently chasing a "ChatGPT moment for robotics". Eka, founded by @pulkitology and @haarnoja
, is taking a different approach to most, and it has already produced an eerie new kind of robot intelligence. (My latest piece for @WIRED
) -
Understanding AI: How ChatGPT and Modern AI Tools Work
By
–
🤖 AI can feel almost magical, but to use it well, we need to understand what it is, how it works, and where its limits really are.
— Bernard Marr (@BernardMarr) 29 avril 2026
In this video, I explain AI in simple terms, including how tools like ChatGPT, Gemini, and Copilot generate responses, why they can sound so… pic.twitter.com/4A5hQAHtwhAI can feel almost magical, but to use it well, we need to understand what it is, how it works, and where its limits really are. In this video, I explain AI in simple terms, including how tools like ChatGPT, Gemini, and Copilot generate responses, why they can sound so
-
BAIR Faculty Elected to ISCB 2026 Fellows Class
By
–
Congratulations to BAIR faculty @jlistgarten and @yun_s_song who have been elected to the International Society for Computational Biology (
@iscb
) 2026 class of Fellows! https://
iscb.org/about-iscb/soc
iety-communications/announcements/april-21-2026-iscb-announces-the-2026-class-of-fellows
… -
Ten Most Popular Machine Learning Models Explained
By
–
10 Most Popular #MachineLearning Models
by @goyalshaliniuk #ArtificialIntelligence #AI #ML #MI