POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
arXiv preprint arXiv:2603.10101 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
ConSPO introduces a contrastive sequence-level policy optimization that aligns rollout scores with generation likelihoods via length-normalized log-probabilities and an InfoNCE-style group contrast with curriculum margin to outperform GRPO on LLM math reasoning benchmarks.
citing papers explorer
-
Beyond Negative Rollouts: Positive-Only Policy Optimization with Implicit Negative Gradients
POPO uses bounded importance sampling on positive rollouts and a siamese policy network to achieve implicit negative gradients and stable optimization, matching or exceeding GRPO on math benchmarks such as 36.67% on AIME 2025.
-
Revisiting Reinforcement Learning with Verifiable Rewards from a Contrastive Perspective
ConSPO introduces a contrastive sequence-level policy optimization that aligns rollout scores with generation likelihoods via length-normalized log-probabilities and an InfoNCE-style group contrast with curriculum margin to outperform GRPO on LLM math reasoning benchmarks.