Sashrika Pandey

[Sashrika Pandey @sashrika_]

Check out our #ICLR2025 paper on Context Steering, where we present an inference-time method that modulates the level of contextual inference for controllable generation ✍️

Joint work with @_herobotics_ @mariah17schrum @ancadianadragan

[Jasper Lu @lu__jasper]

Sharing my first of hopefully many research blog posts! https://t.co/qPvLWmsghr

This one is the kind of educational blog post I wish I'd had when I started with RL for LLMs.

I tried to make it as open as possible. Every rollout is browsable, the code is open source, and I walk through my entire thought process, from learning rate sweeps to reward shaping.

[Architect Labs @architectlabs]

We gave our AI system a spec, and in under 2 weeks, it designed, verified, and deployed a chip that beats NVIDIA.

It’s built for low-power physical AI workloads. We’re running live inference on >B+ parameter models like Llama, Qwen, and Kimi, serving at 3.4x better perf/watt than NVIDIA Jetson.

From just a specification, our AI system autonomously generated all of the RTL Design, UVM verification, formal proofs, firmware, drivers, and kernels, co-designing the model, software and silicon as one optimization loop.

Better AI can now design better chips to run AI, leading to a loop of recursive-self improvement towards our path to abundant intelligence.

[Isadora White @isadorcw]

The real world is an embodied multi-agent system with natural language communication. What if we had a benchmark and platform to study those challenges?

⛏️Introducing MINDcraft and MineCollab, the 1st platform and benchmark for studying embodied multi-agent LLM collaboration!

[Sashrika Pandey @sashrika_]

Presenting this tomorrow! Come chat about steerability and applications of controllable generations #ICLR2025

🕟 Fri 4/25 3-5:30 PM (Poster session 4)
📍Hall 3 + Hall 2B #297

[Sashrika Pandey @sashrika_]

Check out our #ICLR2025 paper on Context Steering, where we present an inference-time method that modulates the level of contextual inference for controllable generation ✍️

Joint work with @_herobotics_ @mariah17schrum @ancadianadragan

[Sashrika Pandey @sashrika_]

Excited to present this work at #ICLR2025, come say hi!

📕 https://t.co/FVKRquKLSo
💻 https://t.co/7eROH9Q3Dc

[Sashrika Pandey @sashrika_]

Check out our #ICLR2025 paper on Context Steering, where we present an inference-time method that modulates the level of contextual inference for controllable generation ✍️

Joint work with @_herobotics_ @mariah17schrum @ancadianadragan

[Sashrika Pandey @sashrika_]

We demonstrate that CoS can be used across API gated models (no weights necessary!) Check out our paper for more experiments (factuality, content modulation, multi-context steering)!

[Sashrika Pandey @sashrika_]

swing by to chat about controllable generation, personalization & bias mitigation! #NeurIPS2024

sat 12/14
📍11-12 SoLaR | West 121-122
📍2-3:30 Behavioral ML | East Mtg Rm. 19, 20
sun 12/15
📍10-11:15 Interpretable AI | East Ballroom A, B
📍3-5 SafeGenAI | Exhibition Hall A

[Sashrika Pandey @sashrika_]

come say hi at #NeurIPS2024 🍁- I'll be presenting our workshop paper on Context Steering, an inference-time method that modulates the level of contextual inference to increase personalization and decrease bias

paper 👉 https://t.co/xkp6f3xrNK

[Sashrika Pandey @sashrika_]

come say hi at #NeurIPS2024 🍁- I'll be presenting our workshop paper on Context Steering, an inference-time method that modulates the level of contextual inference to increase personalization and decrease bias

paper 👉 https://t.co/xkp6f3xrNK

[Sashrika Pandey @sashrika_]

2-3:30 PM today in Jasmine! Come talk to us about culturally aware agents in gameplay 🤩 #EMNLP2024

[Isadora White @isadorcw]

Excited to share that I will be at #EMNLP2024 presenting our work Communicate to Play with @sashrika_ ! 🎊

🕑Session 12 14:00 - 15:30
📍Jasmine - Lower Terrace Level

Come chat about LLM agents, cultural considerations in NLP, and AI for games!

[Isadora White @isadorcw]

Excited to share that I will be at #EMNLP2024 presenting our work Communicate to Play with @sashrika_ ! 🎊

🕑Session 12 14:00 - 15:30
📍Jasmine - Lower Terrace Level

Come chat about LLM agents, cultural considerations in NLP, and AI for games!

[Isadora White @isadorcw]

Excited to announce that Communicate to Play is now accepted to #EMNLP2024 as a findings paper!!

[Sashrika Pandey @sashrika_]

pov: you’re listening to us explain cross cultural communication via AI agents in gameplay

[Isadora White @isadorcw]

Presenting this today at #ACL2024/#ACL2024NLP 🐘

🕹️10:30-11:00: Wordplay poster session
📢11:15-11:45: C3NLP Lightning Talk (Lotus 3)
🕹️12:40-13:40: Wordplay poster session
🕹️15:30-16:00: Wordplay poster session
✨16:00-17:10: C3NLP poster session

Say hi! 🤩

[Sashrika Pandey @sashrika_]

@YikeeeWang @aclmeeting @shangbinfeng amazing congrats!!

[Isadora White @isadorcw]

(1/n)
Check out our new paper Communicate to Play! 📜 We study cross-cultural communication and pragmatic reasoning in interactive gameplay 😃

Paper: https://t.co/f5HB7hxGeY
Code: https://t.co/F8OlyAcaEY
Talk: https://t.co/2okutp1SDq