arXiv:2606. 27771v1 Announce Type: new Abstract: Reinforcement learning (RL) post-training improves the reward alignment of flow-based generators, but often degrades perceptual quality in ways that are not captured by the reward proxy.
arXiv:2606. 08779v1 Announce Type: new Abstract: Reinforcement Learning (RL) has emerged as a pivotal post-training paradigm, yet it frequently suffers from unpredictable sub-optimum performance or even training collapses.
arXiv:2606. 19162v1 Announce Type: new Abstract: Score- and flow-matching models often rely on preference-based reinforcement learning for two purposes: aligning with subjective preferences and, surprisingly, recovering properties such as visual realism and coherent object structure that matching-based training is intended to learn from the data itself.
The paper introduces Stepwise Marginal Information Gain (MIG), an intrinsic process reward that evaluates how each reasoning step of a large language model (LLM) or vision-language model (VLM) improves the likelihood of the reference answer. MIG rewards only new likelihood maxima, preventing duplicate credit, and is combined with outcome, format, and self‑distillation objectives to guide training. Experiments on eight benchmarks show that this method outperforms outcome‑only reinforcement learning and improves accuracy by up to 4.8 points over binary‑reward training, including a 12.6‑point gain on MathVerse and a 12.9‑point advantage on vision‑language transfer at 7B parameters.
arXiv:2602. 18037v2 Announce Type: replace-cross Abstract: Reinforcement Learning from Human Feedback (RLHF) or Verifiable Rewards (RLVR) are two key steps in the post-training of modern Language Models (LMs).
arXiv:2603. 06957v2 Announce Type: replace-cross Abstract: We study post-training linear autoregressive models with outcome and process rewards.
arXiv:2609. 22041v1 Announce Type: new Abstract: Reinforcement learning is increasingly used to align image generators with reward signals, and Flow-GRPO recently extended this paradigm to flow-matching models by treating the denoising sampler as a stochastic policy that can be optimized from reward feedback.
RAGDiffusion++ advances garment generation by addressing the high‑frequency texture gap that previous retrieval‑augmented models left unresolved. The approach introduces a dual‑image FLUX architecture trained on a large, complex garment dataset, coupled with a new attribute‑aware reward model that guides reinforcement learning to favor realistic high‑frequency patterns. An adversarial‑regularized RL strategy (AR‑GRPO) further prevents artifact exploitation, ensuring the model samples authentic, detailed garment textures.
VideoAWR introduces a decoupled-timestep advantage‑weighted rectification framework for video flow models, separating guided rollouts from the optimization anchor to avoid geometric guidance blow‑up and cold‑start collapse. The method uses an EMA reference policy for guided rollouts while rectifying against an unguided conditional anchor, proving that the centered advantage cancels the anchor in expectation and retains only reward‑correlated covariance. Experiments on Wan2.1‑1.3B and 14B models show stable training and improved visual and motion quality in both CFG‑free and CFG‑enabled settings.
arXiv:2609.00892v1 Announce Type: new Abstract: Rubric-based reinforcement learning decomposes open-ended instructions into prompt-specific, flexible rubrics, making it better suited than reinforceme...
arXiv:2606. 29476v1 Announce Type: cross Abstract: Self-distilled agentic reinforcement learning augments trajectory-level reward with a token-level distillation loss, using as its teacher the same policy conditioned on privileged context.
The paper introduces L2C‑TFM, a reinforcement‑learning framework for cleaning tabular data before feeding it to Tabular Foundation Models (TFMs). It proposes a model‑aware reward that regularizes the Wasserstein distance between cleaned and dirty data, aiming to preserve distributional stability. Experiments on ten OpenML datasets show that while some reward designs fail, the model‑aware reward performs comparably to a random‑forest baseline and improves minority‑class macro‑F1 under class imbalance, and a policy trained on one dataset can transfer to others.
The paper presents an unsupervised multimodal reward model built on a frozen Qwen3‑VL backbone. By adding a linear head to an intermediate layer and training it to rank a web caption higher than a perturbed version of itself, the authors avoid human preference labels and achieve competitive performance—outperforming GPT‑4o on several benchmarks with far fewer parameters and compute. The approach demonstrates that a carefully chosen intermediate layer, simple perturbation-based training, and verifier‑driven fine‑tuning can yield effective vision‑language rewards without costly annotation.