PopuLoRA shows that co-evolving populations of LoRA adapters through cross-evaluated self-play can outperform compute-matched single-agent baselines on multiple code and math reasoning benchmarks.
Machine Learning , volume =
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Training-inference mismatch in separated rollout and optimization stages of LLM RL can independently cause training collapse.
citing papers explorer
-
PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-Play
PopuLoRA shows that co-evolving populations of LoRA adapters through cross-evaluated self-play can outperform compute-matched single-agent baselines on multiple code and math reasoning benchmarks.
-
Diagnosing Training Inference Mismatch in LLM Reinforcement Learning
Training-inference mismatch in separated rollout and optimization stages of LLM RL can independently cause training collapse.