2 papers
cs.LG2026
EvalStop: Using World Feedback to Detect and Correct Reward Overoptimization in Multi-Tenant RLHF Platforms
Guilin Zhang, Chuanyi Sun, Kai Zhao +3
Cloud LLM fine-tuning platforms increasingly serve RLHF workloads, where a learned reward model is optimized as a proxy for human quality. As Gao et al. (2023) showed, this proxy d…
cs.LG2026
When Does Deep RL Beat Calibrated Baselines? A Benchmark Study on Adaptive Resource Control
Guilin Zhang, Chuanyi Sun, Kai Zhao +3
A properly calibrated rule-based autoscaler can beat every one of six mainstream deep reinforcement learning (DRL) algorithms on cost across every workload we test - so when, if ev…