Jianwei Liu
@jianweiliu93.bsky.social
PhD Student at UCL Foundational AI CDT;
Focusing on Legged robotics and GenAI;
Experienced robotist;
Prev Biomedical engineering @ Imperial College London;
https://jianwei-liu93.github.io/
Focusing on Legged robotics and GenAI;
Experienced robotist;
Prev Biomedical engineering @ Imperial College London;
https://jianwei-liu93.github.io/
Pretty impressive new simulation framework - Genesis genesis-embodied-ai.github.io "takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 " genesis-world.readthedocs.io/en/latest/us...
Genesis
genesis-embodied-ai.github.io
December 19, 2024 at 4:29 PM
Pretty impressive new simulation framework - Genesis genesis-embodied-ai.github.io "takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 " genesis-world.readthedocs.io/en/latest/us...
Reposted by Jianwei Liu
Introducing MASt3R-SLAM, the first real-time monocular dense SLAM with MASt3R as a foundation.
Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map.
With @ericdexheimer.bsky.social* @ajdavison.bsky.social (*Equal Contribution)
Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map.
With @ericdexheimer.bsky.social* @ajdavison.bsky.social (*Equal Contribution)
December 16, 2024 at 3:43 PM
Introducing MASt3R-SLAM, the first real-time monocular dense SLAM with MASt3R as a foundation.
Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map.
With @ericdexheimer.bsky.social* @ajdavison.bsky.social (*Equal Contribution)
Easy to use like DUSt3R/MASt3R, from an uncalibrated RGB video it recovers accurate, globally consistent poses & a dense map.
With @ericdexheimer.bsky.social* @ajdavison.bsky.social (*Equal Contribution)
Reposted by Jianwei Liu
Interesting and inspiring keynote talk by Danica Kragic
Learning for Interaction and Interaction for Learning
Missed? Watch it here: https://buff.ly/49SzilH
Learning for Interaction and Interaction for Learning
Missed? Watch it here: https://buff.ly/49SzilH
December 13, 2024 at 5:37 PM
Interesting and inspiring keynote talk by Danica Kragic
Learning for Interaction and Interaction for Learning
Missed? Watch it here: https://buff.ly/49SzilH
Learning for Interaction and Interaction for Learning
Missed? Watch it here: https://buff.ly/49SzilH
Reposted by Jianwei Liu
Have you ever wondered how to train an autoregressive generative transformer on text and raw pixels, without a pretrained visual tokenizer (e.g. VQ-VAE)?
We have been pondering this during summer and developed a new model: JetFormer 🌊🤖
arxiv.org/abs/2411.19722
A thread 👇
1/
We have been pondering this during summer and developed a new model: JetFormer 🌊🤖
arxiv.org/abs/2411.19722
A thread 👇
1/
December 2, 2024 at 4:41 PM
Have you ever wondered how to train an autoregressive generative transformer on text and raw pixels, without a pretrained visual tokenizer (e.g. VQ-VAE)?
We have been pondering this during summer and developed a new model: JetFormer 🌊🤖
arxiv.org/abs/2411.19722
A thread 👇
1/
We have been pondering this during summer and developed a new model: JetFormer 🌊🤖
arxiv.org/abs/2411.19722
A thread 👇
1/
Reposted by Jianwei Liu
Tired of saturated benchmarks? Want scope for a significant leap in capabilities?
🔥 Introducing BALROG: a Benchmark for Agentic LLM and VLM Reasoning On Games!
BALROG is a challenging benchmark for LLM agentic capabilities, designed to stay relevant for years to come.
1/🧵
🔥 Introducing BALROG: a Benchmark for Agentic LLM and VLM Reasoning On Games!
BALROG is a challenging benchmark for LLM agentic capabilities, designed to stay relevant for years to come.
1/🧵
November 21, 2024 at 4:24 PM
Tired of saturated benchmarks? Want scope for a significant leap in capabilities?
🔥 Introducing BALROG: a Benchmark for Agentic LLM and VLM Reasoning On Games!
BALROG is a challenging benchmark for LLM agentic capabilities, designed to stay relevant for years to come.
1/🧵
🔥 Introducing BALROG: a Benchmark for Agentic LLM and VLM Reasoning On Games!
BALROG is a challenging benchmark for LLM agentic capabilities, designed to stay relevant for years to come.
1/🧵
Reposted by Jianwei Liu
This study shows that spike sequences carry information beyond what rates and latency to first spike do:
www.nature.com/articles/s41...
My reactions:
1) Cool to see this in humans.
2) Are people still surprised that spike times carry information beyond rates/first-spike latency?!?!?!
🧠📈 🧪
www.nature.com/articles/s41...
My reactions:
1) Cool to see this in humans.
2) Are people still surprised that spike times carry information beyond rates/first-spike latency?!?!?!
🧠📈 🧪
Neuronal sequences in population bursts encode information in human cortex - Nature
The temporal order of neuronal firing within bursts of population spiking in the human anterior temporal lobe is dependent on the category as well as the identity of the individual stimulus, and this ...
www.nature.com
November 28, 2024 at 4:55 PM
This study shows that spike sequences carry information beyond what rates and latency to first spike do:
www.nature.com/articles/s41...
My reactions:
1) Cool to see this in humans.
2) Are people still surprised that spike times carry information beyond rates/first-spike latency?!?!?!
🧠📈 🧪
www.nature.com/articles/s41...
My reactions:
1) Cool to see this in humans.
2) Are people still surprised that spike times carry information beyond rates/first-spike latency?!?!?!
🧠📈 🧪