TL;DR
Two extensions of RLOO are introduced that target a zero-gradient pathology and are implemented as a small subclass of the project’s existing RLOO trainer; all other components are inherited unchanged so curves are directly comparable across runs.
Online Self-Weighted Fine-Tuning is proposed, a simple method that augments SFT with online, trajectory-level weighting and offers a favorable compute-performance trade-off as a practical approach for fine-tuning small-to-medium LLMs on binary-verifiable reasoning tasks with only 2 online rollouts.
The dominance of Neural Networks (NNs) in RL is partially due to their incremental learning capability, which naturally suits the online, non-stationary nature of self-play training. However, gradient-boosted trees like LightGBM are widely recognised as the state of the art for tabular data in supervised learning, ofte...
The sample efficiency and scalability of RL post-training for video MLLMs and introduces OraRL, a decoupled advantage estimator that scales with model size and data, surpassing its backbone from 0.8B to 9B and GRPO up to 100k prompts.
GFlowRL, a streamlined GFlowNet-style RL algorithm that removes the auxiliary partition network entirely while preserving the reward-distribution-matching objective, is proposed, and is the first GFlowNet-style RL algorithm to scale stably across both dense and sparse architectures.
RSTG (Recovering Learning Signals via Adaptive Teacher Guidance), which applies distillation selectively and precisely where it matters most, and substantially outperforms naive GRPO+OPD.
AdaKP is an online selector that re-chooses each problem's KP subset over the course of RL training, an entropy proxy that scores a KP by the reduction in next-token entropy it induces in a single inexpensive forward pass, with a provable bound on its truncation bias.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
