Building simulations and forecasting @ https://xitter.cf/t.co/MmPudR8o20 | ex - RF @ MSR | IIT Jodhpur
Joined December 2023
- Tweets17
- Following186
- Followers65
- Likes211
Atharva Pandey retweeted
I will be at #CoLM2026 presenting our paper "Reason-Mediated Behavioral Models for Auditing LLM Social Simulators" (led by @cloneofatharva)
Can an LLM give the right answer for the wrong reasons?
As we build simulations for social science, we also need to ask whether simulated reasons align with what people actually say.
In our study, human-stated reasons improve behavioral prediction but the tested LLMs often echo a product’s selling points rather than recover why's behind it.
If you are working on simulations of both human and agent behvaior, let's chat!
🔗 arxiv.org/abs/2607.24649
Looking forward to SF? Well, we have the best local guide to whet your appetite
Prepared by our local chair @jurafsky !
docs.google.com/spreadsheets…
Atharva Pandey retweeted
What's the alternative to UNPROMPTED ADS in AI chat?
Just wanted to get out of the house, saw the event (hosted by @stripe @Cloudflare @AnthropicAI @BrainbaseHQ), & I built one possible answer:
A clearing house for agents where brands bid for placement in consumer-agent recommendations, and the highest bidder doesn't automatically win but the user value does.
Won FIRST place with one hour of experimentation!
Product demo edited by dearly (@usecardboard)
Come with an idea. Leave 8 hours later with a startup.
On Day 271 of the singularity, we’re excited to partner with @AnthropicAI, @Cloudflare, and @stripe to host the first Startup Speedrun Hackathon in San Francisco, featuring $27.5K+ in winner prizes and thousands more in participant credits and perks.
The premise is simple: take a startup idea and get as far as humanly (or otherwise) possible in 8 hours.
Build the product. Deploy the infrastructure. Set up payments. Give your agents tools. Find customers.
Winners will get free incorporation and thousands in cash and compute to kickstart their startup.
Apply here: luma.com/g42o84ln
Atharva Pandey retweeted
Excited to be a part of it!
Introducing the Rajeev Circle Fellows 2026.
A cohort of emerging Indian founders and innovators has been selected for the Rajeev Circle Fellowship 2026, bringing together entrepreneurs building startups across a variety of sectors.
The fellowship brings together India’s next generation of technology and innovation leaders, creating a platform to exchange ideas, build meaningful connections, and contribute to conversations in the global innovation ecosystem.
#RCF2026 #RajeevCircle #RajeevCircleFellowship #MotwaniJadejaFoundation
@ashajadeja @garvitgoel03 @gautam_jajoo @nirajmarathe @ThePranitMehta @dangibuilds @yash_goliya @coolcroptech @GalaxEye @FaclonLabs
I'm starting to think humans are just AI wrappers now.
Asked someone for feedback today. Got a AI answer back. We've come full circle. 😺
Atharva Pandey retweeted
What happens when you compare the distributions of real and simulated user behaviors?
🔍 The gap is large.
We introduce a method to measure this gap and evaluate 24 LLM-based user simulators across coding and writing tasks.
@convai_uiuc @MSFTResearch @berkeley_ai
🧵 1/N
Atharva Pandey retweeted
The better LLMs get at reasoning, the longer their traces get—thousands of tokens, dozens of tool calls. But in law, medicine, and agentic AI, "usually correct" isn't good enough: answers must be verifiably correct.
We built interwhen at @MSFTResearch to make that tractable. And it's now open source.
Across benchmarks, plugging interwhen into an LLM yields:
✅ 100% soundness (with full verifiers)
📈 up to 15% accuracy gain
⚡ ~ 1.5× compute cost
🧵
Hello everyone!
We just released a small teaser of what we have been working on. Check it out, and hmu if you want to talk more about this space.
We believe forecasting and simulations should be accessible at SCALE,
the way everyone uses weather forecasts today!
For the past few months, we have been building towards that VISION.
More details coming soon!! Stay in touch. @KairosityAI
Atharva Pandey retweeted
Great discussion this evening at the Hidden Layers reading group on State-Space Models. Thanks to @SaiMadhavanG (IIIT-B) and @cloneofatharva (Kairosity Co-founder) for leading us through it with their insights!
So cool...
Astitva by Anurag, Prasad and Sagar.
Best Overall Film at Mumbai AI Film Festival
Prize: 12 Lakhs in cash + a trip with the best creators of the world to Japan in 2026
Award sponsored by @morphic
You are an AI agent... But it's @gautam_jajoo with a spreadsheet and three coffees
