works on
language models 1long-term memory 1parameter efficiency 1retrieval 1scaling 1

From the 1 of 21 linked papers with an AI index.

collaborators
Showing cs.AIShow all

1 paper · 1 filter

cs.AI2026

Controlling Exploration-Exploitation in GFlowNets via Markov Chain Perspectives

Lin Chen, Samuel Drapeau, Fanghao Shao +5

Generative Flow Network (GFlowNet) objectives implicitly fix an equal mixing of forward and backward policies, potentially constraining the exploration-exploitation trade-off durin…