arXiv:2407. 09013v2 Announce Type: replace Abstract: The attempt to utilize machine learning in PCG has been made in the past.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.
Single domain generalization (SDG) aims to learn a robust model, which could perform well on many unseen domains while there is only one single domain available for training. One of the promising directions for achieving single-domain generalization is to generate out-of-domain (OOD) training data through data augmentation or image generation.
arXiv:2608. 05248v1 Announce Type: new Abstract: Generating large-scale, freely explorable 3D worlds from open-ended text remains challenging because a system must jointly maintain global spatial coherence, rich local content, and explicit assets suitable for downstream editing and reuse.
The paper "Measuring Human Contribution in AI-Assisted Content Generation" addresses the challenge of determining how much human input influences content produced with generative AI. It proposes an information-theoretic framework that calculates the mutual information between human input and AI output relative to the self-information of the output, thereby quantifying the proportion of human contribution. Experiments across various creative domains show that this measure can distinguish different levels of human involvement in AI-assisted works.
arXiv:2602. 03300v2 Announce Type: replace-cross Abstract: In this work, we aim to develop effective data synthesis techniques that autonomously synthesize multimodal training data for enhancing MLLMs in solving complex real-world tasks.
arXiv:2509. 09960v2 Announce Type: replace-cross Abstract: Synthetic tabular data generation is increasingly essential in machine learning, supporting downstream applications when real-world, high-quality tabular data is insufficient.
RubricRM introduces a pairwise generative reward modeling framework that generates an input‑specific rubric—comprising evaluation dimensions, weights, and scoring criteria—to score candidate images. The method is trained in two stages: supervised fine‑tuning to learn the rubric‑based scoring paradigm and GRPO to refine dimension‑level rewards. Experiments on text‑to‑image generation and instruction‑based image editing benchmarks demonstrate that RubricRM outperforms existing specialized reward models and competes with strong proprietary MLLM judges while using smaller backbones.