Aakash Yadav

Ph.D. @UWMadison | prev @honda research institute @TAMU @iit_tirupati | HRI, HAI, Affective computing, robotics | views are my own

[Aakash Yadav @nimrobotics]

Super excited to lead HFES UW-Madison Student Chapter this year! 🚀

[HFES UW-Madison @HFESMadison]

✨ Meet Our Leadership Team – Fall 2025 ✨

We’re excited to introduce the new HFES UW-Madison executive board!

This amazing team will be leading our efforts in building community, hosting events, and advancing human factors at UW. Stay tuned for upcoming activities! 🚀

[AI at Meta @AIatMeta]

Today is the start of a new era of natively multimodal AI innovation.

Today, we’re introducing the first Llama 4 models: Llama 4 Scout and Llama 4 Maverick — our most advanced models yet and the best in their class for multimodality.

Llama 4 Scout
• 17B-active-parameter model with 16 experts.
• Industry-leading context window of 10M tokens.
• Outperforms Gemma 3, Gemini 2.0 Flash-Lite and Mistral 3.1 across a broad range of widely accepted benchmarks.

Llama 4 Maverick
• 17B-active-parameter model with 128 experts.
• Best-in-class image grounding with the ability to align user prompts with relevant visual concepts and anchor model responses to regions in the image.
• Outperforms GPT-4o and Gemini 2.0 Flash across a broad range of widely accepted benchmarks.
• Achieves comparable results to DeepSeek v3 on reasoning and coding — at half the active parameters.
• Unparalleled performance-to-cost ratio with a chat version scoring ELO of 1417 on LMArena.

These models are our best yet thanks to distillation from Llama 4 Behemoth, our most powerful model yet. Llama 4 Behemoth is still in training and is currently seeing results that outperform GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on STEM-focused benchmarks. We’re excited to share more details about it even while it’s still in flight.

Read more about the first Llama 4 models, including training and benchmarks ➡️ https://t.co/9G3QgVdCkB
Download Llama 4 ➡️ https://t.co/eVomRvEr0w

[Aakash Yadav @nimrobotics]

What’s more fun than learning things by doing! Below are some instruments for one of the many many interactive class on intro to human factors I have been teaching this semester

[Aakash Yadav @nimrobotics]

Grateful for the amazing mentorship and support from @ranjanamehta and a lot of other people!! 🏆

#isye #hri #robotics

[Andrej Karpathy @karpathy]

The YouTube video I want to watch is any highly rated, 1hr long, information dense lecture on anything esoteric and the algorithm just doesn’t get it. It’s too content-driven and too narrow-minded

[Aakash Yadav @nimrobotics]

Just delivered my first lecture! Great start to the fall semester!!

[Aakash Yadav @nimrobotics]

Update: I’m back in Madison after completing my research intern at Honda research institute in Bay Area. Lots of learnings, food and hikes

[Aakash Yadav @nimrobotics]

It’s 2024 and excel cannot open two files with same name 😒

(LibreOffice comes to the rescue)

[Andrej Karpathy @karpathy]

found in the source code

[Aakash Yadav @nimrobotics]

this year I went from houseplants to a full fledged garden and this has been the most rewarding and serene experience of all time! some pictures showing the first harvest. The fun part: almost everything was planted from a seed 🧑🏼‍🌾

[Aakash Yadav @nimrobotics]

@philartatVT Congratulations!! 🎊🎉

[Aakash Yadav @nimrobotics]

I’m spending this summer as a research intern at Honda Research Institute working with the Human Centered Intelligence team! 🚀

#honda #intern #research #hondaresearchinstitute

[Pushmeet Kohli @pushmeet]

Happy to announce #AlphaFold 3 from @GoogleDeepMind and @IsomorphicLabs. Our new AlphaFold Server illuminates the molecular universe in cells (from proteins to RNA, DNA, ligands and beyond) transforming our understanding of biology.

Find out more here ⬇️
https://t.co/L8LUPtNfOE

[ACM Digital Library @ACMDL]

Check out "Beyond Dyadic Interactions: Assessing Trust Networks in Multi-Human-Robot Teams," by Aakash Yadav, Ranjana Mehta, an Honorable Mention: Late Breaking Report from HRI '24: ACM/IEEE International Conference on Human-Robot Interaction: https://t.co/QvFv73woy4

[Aakash Yadav @nimrobotics]

My undergrad mentee Aarav Gupta @WisconsinCS presented his research on eye gaze synchrony in mHRT at @UWMadison Undergraduate Symposium!!
Aarav has been doing some fantastic work this semester at @neuro_ergo @uwisye

#hri #ugresearch #mentorship

[ollama @ollama]

.@Meta Llama 3 - The most capable openly available LLM to date!

ollama run llama3

https://t.co/T1XAqEC299

If you have pull the llama 3 model prior to this post, please update the model using `ollama pull`.

[Aakash Yadav @nimrobotics]

was trying to transcribe some real-world (noisy) data and open source wins again by a large margin

[Andrej Karpathy @karpathy]

Returning from an experimental ~2 week detox from the internet. Main takeaway is that I didn't realize how unsettled the mind can get when over-stimulating on problems/information (like a stirred liquid), and ~2 weeks is enough to settle into a lot more zen state.

I'm struck by how an over-stimulated brain automatically keeps bubbling up problems into consciousness, creating a state of persistent anxiety and nervousness. After some time, in the settled state, this activity just... stops. You can sit down and your brain doesn't immediately go into some kind of problem solving overdrive, it just stays silent. Nothing happens.

I'm sure this could read a bit duh to many, but I haven't been to this subset of "brain dynamics" state space in I think a very long time and it is comforting to know that 1) it exists, and 2) you can visit, if you like, but the journey there takes a few weeks.

Anyway, where were we :D

[UW Ind & Sys Eng @uwisye]

#PhD students Aakash Yadav, Yinsu Zhang & Sarah Hopko were on hand at the 2024 International Conference on #Human-Robot Interaction to present their work. Aakash's paper, w/Prof. @ranjanamehta, received honorable mention under "Best Late Breaking Report". #HRI2024 @UWMadEngr

[Aakash Yadav @nimrobotics]

Also, squeezed in a tiny but beautiful hike!! 🥾

[Aakash Yadav @nimrobotics]

Had a fantastic time presenting my work on multi-human-robot teaming at the 2024 ACM/IEEE International Conference on Human-Robot Interaction (HRI 2024) 🤖 in Boulder. w/w @YinsuZ @ranjanamehta

Learn more: https://t.co/94vfPvYoZQ

@uwisye @neuro_ergo #ACM #IEEE #HRI2024