LLaDA2.0-Uni Unifying Multimodal Understanding and Generation with Diffusion Large Language Model paper: https://
huggingface.co/papers/2604.20
796
…
AI
-
LLaDA2.0-Uni: Unified Multimodal Understanding and Generation with Diffusion
By
–
-
Claude vs Other AI: Emotional Behavior and Tone Comparison
By
–
I did. Still behaves too emotional. Maybe my instructions aren’t good enough but Claude has this precise, serious tone from the beginning. Still prefer Claude a lot more.
-
Chinese Humanoid Robots Outrun Human Marathon World Record
By
–
Breaking news! Chinese humanoids are now officially faster than humans. 🤯
— Andreas Klinger 🦾 (@andreasklinger) 23 avril 2026
The Beijing E-Town Half Marathon just happened. Humans and humanoids running in parallel. Over 300 robots competed. And the winner, Honor Lightning, beat the human world record holder by seven minutes.… pic.twitter.com/i9PHDdH5ehBreaking news! Chinese humanoids are now officially faster than humans. The Beijing E-Town Half Marathon just happened. Humans and humanoids running in parallel. Over 300 robots competed. And the winner, Honor Lightning, beat the human world record holder by seven minutes.
-
BytePlus Coding Plan: AI-Powered Programming Platform Launch
By
–
try it here:
→ https://
byteplus.com/en/activity/co
dingplan?utm_source=PR_KOL&utm_medium=kol&utm_campaign=BP-Global-Codingplan&ArkClaw-publish-Q2APRFY26&utm_term=Aria_Westcott&utm_content=codingplan
… -
GLM 5.1 Local Model Matches Opus 4.6 Performance
By
–
🚨 WE OFFICIALLY HAVE GLM 5.1 🚨
— Charly Wargnier (@DataChaz) 23 avril 2026
A local model as good as Opus 4.6 (in some evals), running right on a 256GB Mac Studio 🤯
WILD stats → 1,432 tokens / 1:17s / 18.47 t/s
..and it absolutely crushes SWE-Bench Pro!@BytePlusGlobal just bundled it with 5 other elite models 👀↓ https://t.co/cXK5Km54tS pic.twitter.com/2pTRXPge3sWE OFFICIALLY HAVE GLM 5.1 A local model as good as Opus 4.6 (in some evals), running right on a 256GB Mac Studio WILD stats → 1,432 tokens / 1:17s / 18.47 t/s ..and it absolutely crushes SWE-Bench Pro! @BytePlusGlobal just bundled it with 5 other elite models ↓
-
Workday Rippling Slack fail data access security test
By
–
Workday, Rippling, and Slack flunk data access test, claims Fivetran https://
go.theregister.com/feed/www.there
gister.com/2026/04/22/workday_rippling_slack_flunk_fivetran_data_access_test/?utm_source=dlvr.it&utm_medium=twitter
… -
GPT 5.5 needs serious communication style like Claude
By
–
I hope that spud (gpt 5.5) will also get a new, more serious way of communicating. What I like about Claude is this seriousness, this concise and confident communication. I dislike ChatGPT's "chattyness" in everyday use, this forced "emotional understanding," even when you adjust
-
ICLR 2026 Outstanding Papers Selection Procedure Announced
By
–
More details on the recognized papers and the selection procedure here: https://
blog.iclr.cc/2026/04/23/ann
ouncing-the-iclr-2026-outstanding-papers/
… -
ICLR 2026 Outstanding Paper Award Winners Announced
By
–
Congratulations to the winners of the #ICLR2026 Outstanding Paper Award! It was an honour to chair the committee, the entire team did a great job. We recognized two outstanding papers and one honorable mention. Check out the blog post for more: https://
blog.iclr.cc/2026/04/23/ann
ouncing-the-iclr-2026-outstanding-papers/
…