fitness.mp4

More of this tonight! 👊

Morning🌻the bear woke up 🐻

😂😂 Sunday special 😂😂

10 reps of 45 lbs. My morning workout preparing... to fight crime in DC 🤣🤣🤣🇺🇸show more

We’re approaching October 7th soon. Last year in London... we commemorated the tragic loss of our angels who were murdered that day. I always bring both the Iranian and Israeli flags to these types of gatherings, rallies and protests. Barely 3 months after that, the January Genocide began in Iran. I’m still stuck in 2025. 2026 never began for me and for millions of Iranians. I pray this upcoming Rosh Hashanah will bring us out of this years long darkness and towards light and freedom. May our people live in peace and dignity. It’s highly overdue now.show more

me waking up for work at 6am after going... to bed at 5amshow more

"you attract what you fear" me:

Happy Sunday 🥰

Does LLM really need to be a helpful assistant... all the time? No. If you want to simulate people, “perfectly helpful” could be the wrong objective. Meet OdysSim, a journey toward LLMs beyond assistants, as behavioral foundation models (10B tokens of real human behavior; 23 sim benchmarks, finally in one place. new open models: outperform or on par with GPT-5.5, Gemini 3.1, or Claude Opus 4.7 in many behavior-sim dimensions). Human behavior simulation is becoming essential. Agent evaluation needs realistic users before real users show up. Medical and classroom training need realistic patients and students. Social science needs synthetic participants at scale. But real people are not ideal assistants. Real patients panic or ignore good advice. Real students misunderstand. Real customers are vague, picky, impatient, or simply leave. Human behavior is messy, diverse, and often imperfect. Frontier LLMs are getting better at math, code, and long-horizon tasks. They are NOT getting better at simulating human behavior. If anything, they drift the other way: more assistant-ish, more homogeneous, fewer of the errors and quirks real humans show. This is no accident. The whole pipeline is built for helpfulness and task success, not behavioral realism. And you can't prompt your way out of that. So we rethink the recipe from scratch and release: 🧠 The OdysSim corpus: 21.4M real human interactions (~10B tokens) from 62 sources, every conversation retrofitted with social grounding (who is talking, and why) 📏 SOUL-Index: 23 human-behavior benchmarks unified into one suite across 5 axes 🤖 OSim-8B: open weights; tops more SOUL-Index benchmarks than any frontier model, acts more like a real user than any of them on τ-bench (nearly matching real humans in the reaction dimension), and writes far more human-like text along the way.show more

"Real GDP growth is 2.6%, not 7.8%. The government... is good at revising the base-year data so that it can project GDP growth as high as 7.8%." — Subash Chandra, Ex Finance Secretaryshow more

Do u like my outfit?

Taehyung saying "dododo" made jungkook laugh in the middle... of singing 😭😭😭show more

During Eid peak… Dubai hotels hit 80–90% occupancy 🔥... Not luck. A city built to attract the world, all year round.show more

bittersweet mingyu in the year 2026??!!

treadmill running is boring.🥲

Mintzy is torn between wanting to get into the... concert as fast as possible and walking with (redacted) Currently the concert is winning the battle.show more

Ben Shapiro can be pretty funny at times.

Cats the way uhuh uhuh...

Jonah from Tomb Raider has a quote that I... love so much. "I like this world. It's not perfect but everything I love now is in it."show more

that nigga Ar Ab used to be going too... federal ctfu 😂😂😂 .. peep main man face in the redshow more