Riashat Islam (@riashatislam) / X

Riashat Islam

615 posts

Riashat Islam

@riashatislam

Research Scientist @ms_aifrontiers @MSFTResearch NYC; Ex @HUMAIN @DreamFoldAI PhD @Mila_Quebec, intern @MSRNYC @AppleMLR; RL, Reasoning and LLMs; WorldModels

New York, USA

Joined November 2016

Pinned
Riashat Islam
@riashatislam
Apr 25, 2023
Excited to share that Agent-Controller representations for offline RL in presence of rich exogenous information is now accepted at #ICML2023 (arxiv.org/abs/2211.00164) This is a follow-up of our recent work on latent state discovery (#TMLR'23) arxiv.org/abs/2207.08229
arxiv.org
Agent-Controller Representations: Principled Offline RL with Rich...
Learning to control an agent from data collected offline in a rich pixel-based visual observation space is vital for real-world applications of reinforcement learning (RL). A major challenge in...
Riashat Islam
@riashatislam
Aug 1
After Energy-Based Transformers, @AlexiGlad comes back with another really cool work!
Alexi Gladstone
@AlexiGlad
Jul 31
We discovered a third pretraining axis beyond parameters and data: exploration. Scaling exploration monotonically improves existing models across images/video/language, and unlocks end-to-end generation. In the simplest case, it's just a for loop. Introducing Explorative
Riashat Islam
@riashatislam
Jul 4
Check out our poster on h1 and find @sumeetrm for a chat!
Sumeet Motwani
@sumeetrm
Jul 3
I'll be presenting 3 papers at ICML 2026🫡 h1 (Spotlight) trains models to reason over longer horizons using curriculum RL over composed short-horizon data. This allows models to generalize to harder tasks and improves performance even at high pass@k. LongCoT isolates and
Riashat Islam
@riashatislam
Jun 28
This is really cool work - you know this work has been thought through when you see @EfroniYonathan on it! Congrats to @Ankur_Samanta_ for the hard work in pulling it off too!
Ankur Samanta
@Ankur_Samanta_
Jun 22
🚀New work on credit assignment in multi-step reasoning RL post-training🚀 Introducing Self-Reset Policy Optimization (SRPO): i) localize the first wrong reasoning step, ii) reset to that step, iii) learn from counterfactual continuations from there – no external supervision.🧵
Riashat Islam
@riashatislam
Jun 6
A huge loss for the RL, Control and Optimization communities. A legend passed way.
Subbarao Kambhampati (కంభంపాటి సుబ్బారావు)
@rao2z
Jun 5
Deeply saddened at the passing of my dear colleague, Dimitri Bertsekas. Everyone in RL, OR and control theory already knows of his monumental contributions. Over the past seven years, we at @SCAI_ASU also got to know him as an unwaveringly kind and gracious man of science. He