Twitxy Saved

Haozhe Jiang

@erichzjiang

PhD student at @berkeley_ai. Previously @Tsinghua_IIIS
astro-eric.github.io Joined April 2023
301 Following 896 Followers 59 Posts

Latest posts 5

Haozhe Jiang @erichzjiang
LLM makes generation cheap. Agent makes iteration cheap. In RPG, we build an agentic robot system leveraging offline dataset and iterate on reconstructed tasks, and it could accomplish real world manipulation with really low cost!
Yen-Jen Wang Verified account @wangyenjen Oct 7
Practice makes perfect. A truly agentic robot shouldn’t just know what to do. It should also figure out how to get better.

What if a robot could watch a demonstration, figure out what it needs to practice, and learn how to get better through practice?

Introducing RPG: Guided Self-Improvement for Embodied Agents, a real-to-sim-to-real pipeline that turns real-world demonstrations into targeted practice.

RPG reconstructs practice tasks from demonstrations, uses demo videos to guide skill improvement, and continually updates a shared skill system from failures.

On real robots, RPG achieves 30/30 successes across 3 tasks. In simulation, it reaches 95.0% success across 22 manipulation tasks after 15 rounds of self-improvement.

Demo → Reconstruct → Guided Practice → Self-Improve → Go Real
View quoted post on X
2 replies 2 reposts 14 likes 818 views View on X
Haozhe Jiang @erichzjiang
Unified models unlocks fundamentally new capabilities than LLMs or VLMs. It was super fun studying how generation and understanding correlate, breaking down these capabilities to fine-grained taxonomy, and improve training!
Jiaxin Ge @ CoRL Verified account @jiaxin_ge_ Oct 5
Generation 𝗖𝗔𝗡 improve understanding!

With the right recipe, we show that training on generation data can help understanding tasks.

Introducing OmniTaskonomy: our year-long study of when and how generation helps understanding.
View quoted post on X
0 replies 1 reposts 12 likes 860 views View on X
Haozhe Jiang @erichzjiang
Why aren’t Diffusion Language Model smart yet? Lacking stable post training is a major bottleneck!

Meet DiPOD: the tripod for diffusion model post-training.

DiPOD boosts accuracy across reasoning tasks, with Sudoku jumping from 22% to 97%, through a one-line code change.

🧵1/5
0:30
13 replies 74 reposts 591 likes 140K views View on X