3 sources
One of the most important newer names to track in alignment-flavored language-model work because he sits directly on the line from DPO into newer attempts to turn language models into better optimizers and agents.
Researcher Profile
Editor reviewedArchit Sharma
Direct preference optimization (DPO)
Stanford researcher on preference optimization and autonomous RL
A useful person to follow for the bridge between reinforcement-learning instincts and later alignment methods like DPO, especially where preference optimization is treated as a core learning problem rather than a bolt-on finetuning trick.
Organizations
Stanford UniversityUniversity of Iowa
About This Page
This profile is meant to help you get oriented quickly: why this researcher matters, what to read first, and where to explore next.
Last reviewed
March 18, 2026
Official And External Links
Known For
The ideas, systems, and research directions that make this person worth knowing.
01
Direct preference optimization
02
Autonomous reinforcement learning
03
Evaluating AI feedback for alignment
04
Direct preference optimization (DPO)
05
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
06
DPO
Start Here
Canonical papers, project pages, or repositories that anchor this profile.
Signature Works
Additional papers, projects, or repositories that help flesh out the profile.
Supporting Sources
Additional links that help verify and flesh out this profile.
Related Researchers
People worth exploring next because they share topics, labs, or source material with this profile.
3 sources
A high-signal researcher for the probabilistic and generative-modeling side of modern AI, and an important bridge into the Stanford preference-optimization cluster that helped make DPO mainstream.
Start HereStefano Ermon at Stanford
3 sources
One of the clearest people to follow for the overlap between modern robotics, meta-learning, and preference-optimization-era alignment research.
Start HereChelsea Finn at Stanford HAI
3 sources
A high-signal name for the current alignment toolkit, especially if you want to understand how preference optimization connects back to broader language-model adaptation work.
5 sources
A strong person to follow for how Anthropic moved from assistant training into more explicit evaluation work around model behavior, red-teaming, and chain-of-thought faithfulness.
5 sources
Worth following for the evaluation side of Anthropic’s alignment program, especially where model-written tests and public-input methods become practical tooling rather than just ideas.