
Ishita Mediratta
@ishitamed
Followers
691
Following
10K
Media
183
Statuses
2K
1/ .🤔 Curious about the generalization power of offline RL?. In our latest study, we dive deep into testing how well BC, Offline RL, and Sequence Modeling algorithms fare when faced with new Procgen levels or WebShop instructions . Paper:
2
15
102
RT @holynski_: Another one. Already a powerful painting, but moving around it yourself gives a totally different feeling. Jacques Louis Da….
0
317
0
If you want to build AI Research Agents 🤖 with extremely talented researchers and love NYC 🗽, do apply! 👇.
Hiring! We're looking to fill contractor Research Engineer roles in New York City to work with us in FAIR on AI Research Agents. If that sounds fun, please fill out the expression of interest here:
0
0
4
RT @_jasonwei: Becoming an RL diehard in the past year and thinking about RL for most of my waking hours inadvertently taught me an importa….
0
338
0
RT @kyutai_labs: Talk to 🔊, the most modular voice AI around. Empower any text LLM with voice, instantly, by wrappi….
0
265
0
RT @alex_h_miller: Hiring! Do you love the idea of waiting extra long for training / evals because your model trains more models? ⏳ Does fr….
0
8
0
Eagerly waiting for the day @NotebookLM releases an iOS app! Learning from new papers everyday has become so much easier by listening to those 2 fantastic AI hosts! 🙌.
0
0
2
RT @_samvelyan: Incredible keynote by @_rockt! Some might fear that open-endedness is a disease for AI safety; I say it might be the remedy….
0
10
0
Really like this slide from @j_foerst ‘s talk today at World Modelling Worshop!. Btw, if ACCEL was for UED, is it right to call an autonomous AI scientist “ACCELERATE”? What say @jparkerholder 😉
2
0
25
RT @robertarail: I’ll be giving a talk at 2:20pm today at the @ICBINBWorkshop #ICLR2025 titled “Automating Scientific Discovery: How Far Ar….
0
14
0
RT @astonzhangAZ: Our Llama 4’s industry leading 10M+ multimodal context length (20+ hours of video) has been a wild ride. The iRoPE archit….
0
136
0
RT @sang_yun_lee: How would gpt-4o image generation work? Speculation:. - gpt-4o generates visual tokens, and the diffusion decoder decodes….
0
127
0
RT @nick11roberts: 📉📉NEW SCALING LAW PHENOMENON 📉📉 . We find that knowledge and reasoning exhibit different scaling behaviors! . Super exci….
0
172
0