Search

Rohan Surana

Reasoning Pathologies in Large Language Models: A Diagnostic Perspective
The Geometry of Agent Skills: Non-Commutative Composition in Representation Space
Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering
RRPO: Reference-Relative Policy Optimization with Stratified Conditional Rollouts
Skill-R1: Agent Skill Evolution via Reinforcement Learning
Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.