LLM hidden states contain a sparse reward subsystem consisting of value neurons that predict state value and dopamine neurons that encode step-level temporal difference errors.
LaSeR: Reinforcement Learning with Last-Token Self-Rewarding
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
OGER adds an auxiliary exploration reward built from offline trajectories and model entropy to hybrid RL training, yielding gains on math reasoning benchmarks and out-of-domain generalization.
citing papers explorer
-
Sparse Reward Subsystem in Large Language Models
LLM hidden states contain a sparse reward subsystem consisting of value neurons that predict state value and dopamine neurons that encode step-level temporal difference errors.
-
OGER: A Robust Offline-Guided Exploration Reward for Hybrid Reinforcement Learning
OGER adds an auxiliary exploration reward built from offline trajectories and model entropy to hybrid RL training, yielding gains on math reasoning benchmarks and out-of-domain generalization.