Aviral Kumar
Assistant Professor
Develops reinforcement learning methods, including offline RL and training large models from feedback and interaction.
Aviral Kumar's lab at CMU works on reinforcement learning and large language models. 8 papers since 2023, 4 at top venues, mostly NeurIPS and CoRL. 4 PhD students. Applications close .
At a glance
- papers, 2023–2026
- 2/ year
- at T1 venues
- 50%
- PhD students
- 4
- MS students
- 0
- alumni listed
- —
Papers by yearfrom the lab homepage, 2023–2026
- 0 papers2026 · T1 0 · collecting
- 0 papers2025 · T1 0
- 8 papers2024 · T1 4
- 0 papers2023 · T1 0
T1 venuesOtherVenues: NeurIPS, CoRL
Research focuspaper titles that use each topic's words
- Reinforcement learning5 papers
- Large language models5 papers
- offline1 paper
- develops reinforcement0 papers
- feedback0 papers
Paper list
From the lab homepage.
20248 papers
- International Conference on Machine Learning (ICML)ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL
- NeurIPS (T1)DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning
- NeurIPS (T1)Is Value Learning Really the Main Bottleneck in Offline RL?
- International Conference on Machine Learning (ICML)Preference Fine-Tuning of LLMs Must Use Suboptimal, On-Policy Data
- NeurIPS ; ICML Workshop on LLMs and CognitionRecursive Introspection: Teaching Language Model Agents How to Self-Improve
- NeurIPS (T1)RL on Incorrect Synthetic Data Enhances the Efficiency of LLM Math Reasoning by Eight-Fold
- CoRL (T1)Steering Your Generalists: Improving Robotic Foundation Models via Value Guidance
- International Conference on Machine Learning (ICML)Stop Regressing: Training Value Functions via Classification for Scalable Deep RL
Where alumni went
Alumni are not listed on the lab homepage.
Apply to CMUFall 2027
CMU SCS Computer Science PhD
- Round
- Single round
- Opens
- GRE
- See program page
- English
- TOEFL / IELTS / Duolingo (F-1/J-1 non-native speakers)
- Fee
- USD 100 (USD 80 by early deadline)
Confirmed on the program page.
Similar labsnearest by research summary
- GBUCL
Peter Orbanz
- Learning theory & optimization
- USNYU
Keith Ross
- Reinforcement learning
- GBUCL
Leena Chennuru Vankadara
- Learning theory & optimization
- USUW
Kevin Jamieson
- Reinforcement learning
- USUC Berkeley
Bruno Olshausen
- Computational neuroscience & cognition
- USNYU
Rumi Chunara
- Trustworthy & safe ML
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- The alumni list could not be found on the lab site.
- The lab site and the paper record do not agree.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Not collected
Tiers are venue classifications, not lab rankings.
CMUcloses