S3AC adds a stabilized actor-critic for LLM reasoning
The paper reports gains from 1.5B to 397B parameters and stable single-rollout training at 35B.
Stories per day · 30 days
4 stories, newest first · one card per story
Stories tagged #actor-critic
Paper analyzes continuous-time policy gradients from discrete data
The model-free actor-critic methods use PhiBE and are tested on a linear-quadratic regulator and a wheeled robot.
PACT: From Credit Assignment to Critic Alignment
PACT realigns actor-critic training and beats GRPO and PPO on math reasoning and SWE-bench Verified.
CARE-VI: Conservative Adaptive Reliability Estimation for Value Improvement in Off-Policy Actor-Critic Learning
CARE-VI introduces CARS, SEVA, and DARE components for reliability-aware value targets, achieving highest mean returns across twelve SAC, TD3, and TD7 settings.