Martin Klissarov
@martinklissarov.bsky.social
research @ Google DeepMind
Pinned
As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing.
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing.
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
June 27, 2025 at 8:16 PM
As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing.
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
But how do we discover such temporal structure?
Hierarchical RL provides a natural formalism-yet many questions remain open.
Here's our overview of the field🧵
Reposted by Martin Klissarov
Our team in London is hiring a research scientist! If you want to come work with a wonderful group of researchers on investigating the frontiers of autonomous open-ended agents that help humans be better at doing things we love, come have a look. Link in post below 👇
March 18, 2025 at 4:01 PM
Our team in London is hiring a research scientist! If you want to come work with a wonderful group of researchers on investigating the frontiers of autonomous open-ended agents that help humans be better at doing things we love, come have a look. Link in post below 👇
Reposted by Martin Klissarov
Our paper showing that LMs benefit from human-like abstractions for code synthesis was accepted to ICLR! 🇸🇬
We show that order matters in code gen. -- casting code synthesis as a sequential edit problem by preprocessing examples in SFT data improves LM test-time scaling laws
We show that order matters in code gen. -- casting code synthesis as a sequential edit problem by preprocessing examples in SFT data improves LM test-time scaling laws
February 12, 2025 at 8:08 PM
Our paper showing that LMs benefit from human-like abstractions for code synthesis was accepted to ICLR! 🇸🇬
We show that order matters in code gen. -- casting code synthesis as a sequential edit problem by preprocessing examples in SFT data improves LM test-time scaling laws
We show that order matters in code gen. -- casting code synthesis as a sequential edit problem by preprocessing examples in SFT data improves LM test-time scaling laws
Can AI agents adapt zero-shot, to complex multi-step language instructions in open-ended environments?
We present MaestroMotif, a method for skill design that produces highly capable and steerable hierarchical agents.
Paper: arxiv.org/abs/2412.08542
Code: github.com/mklissa/maestromotif
We present MaestroMotif, a method for skill design that produces highly capable and steerable hierarchical agents.
Paper: arxiv.org/abs/2412.08542
Code: github.com/mklissa/maestromotif
February 4, 2025 at 7:22 PM
Can AI agents adapt zero-shot, to complex multi-step language instructions in open-ended environments?
We present MaestroMotif, a method for skill design that produces highly capable and steerable hierarchical agents.
Paper: arxiv.org/abs/2412.08542
Code: github.com/mklissa/maestromotif
We present MaestroMotif, a method for skill design that produces highly capable and steerable hierarchical agents.
Paper: arxiv.org/abs/2412.08542
Code: github.com/mklissa/maestromotif
Reposted by Martin Klissarov
Our paper on AI feedback was accepted to #ICLR2025 as a poster. Great work by @martinklissarov.bsky.social , @bmazoure.bsky.social , and Alex Toshev
arxiv.org/abs/2410.05656
arxiv.org/abs/2410.05656
On the Modeling Capabilities of Large Language Models for Sequential Decision Making
Large pretrained models are showing increasingly better performance in reasoning and planning tasks across different modalities, opening the possibility to leverage them for complex sequential decisio...
arxiv.org
January 22, 2025 at 5:19 PM
Our paper on AI feedback was accepted to #ICLR2025 as a poster. Great work by @martinklissarov.bsky.social , @bmazoure.bsky.social , and Alex Toshev
arxiv.org/abs/2410.05656
arxiv.org/abs/2410.05656
Reposted by Martin Klissarov
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.
December 4, 2024 at 4:01 PM
Introducing 🧞Genie 2 🧞 - our most capable large-scale foundation world model, which can generate a diverse array of consistent worlds, playable for up to a minute. We believe Genie 2 could unlock the next wave of capabilities for embodied agents 🧠.