markov chains
Mathematical systems that undergo transitions from one state to another in a state space, with the property that the future state depends only on the current state, employed in various AI applications such as probabilistic modeling and reinforcement learning.
- BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning
- Cross-fluctuation phase transitions reveal sampling dynamics in diffusion models
- Distances for Markov chains from sample streams
- Inv-Entropy: A Fully Probabilistic Framework for Uncertainty Quantification in Language Models
- Safety Depth in Large Language Models: A Markov Chain Perspective
- Shift Before You Learn: Enabling Low-Rank Representations in Reinforcement Learning