Scholar

Aviral Kumar

Google Scholar ID: zBUwaGkAAAAJ

Carnegie Mellon University

AIReinforcement Learning

Homepage↗Google Scholar↗

Citations & Impact

All-time

Citations

25,605

H-index

50

i10-index

70

Publications

20

Co-authors

8

list available

Contact

No contact links provided.

Publications

29 items

QED-Nano: Teaching a Tiny Model to Prove Hard Theorems

2026

Cited

0

IsoCompute Playbook: Optimally Scaling Sampling Compute for LLM RL

2026

Cited

0

What Does Flow Matching Bring To TD Learning?

2026

Cited

0

BPP: Long-Context Robot Imitation Learning by Focusing on Key History Frames

2026

Cited

0

Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL

2026

Cited

0

POPE: Learning to Reason on Hard Problems via Privileged On-Policy Exploration

2026

Cited

1

InT: Self-Proposed Interventions Enable Credit Assignment in LLM Reasoning

2026

Cited

1

TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks

2026

Cited

0

Resume (English only)

Co-authors

8 total

UC Berkeley, Physical Intelligence

Google DeepMind

Stanford University, Physical Intelligence

Rishabh Agarwal

Meta, ex DeepMind, Google Brain

Stanford University

Google DeepMind

Google DeepMind