← Back to all news
arXiv AI By Christian Greisinger, Steffen Eger

TikZilla: Scaling Text-to-TikZ with High-Quality Data and Reinforcement Learning

Read the original on arXiv AI →

TikZilla is a new approach to generating TikZ code from textual descriptions, built on a larger, higher‑quality dataset called DaTikZ‑V4 that includes LLM‑generated figure descriptions. The method uses a two‑stage pipeline: supervised fine‑tuning of small Qwen models (3B and 8B) followed by reinforcement learning with an image encoder that provides semantically faithful reward signals. Human evaluations show that TikZilla outperforms its base models by 1.5–2 points on a 5‑point scale, beats GPT‑4o by 0.5 points, and matches GPT‑5 in image‑based tests while remaining much smaller.

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.

Related stories

arXiv Computer Vision
Aug 25

VinciCoder: Unifying Multimodal Code Generation via Coarse-to-fine Visual Reinforcement Learning

VinciCoder is a unified framework for multimodal code generation that addresses the limitations of single-task models by training on a large-scale curated corpus of 1.3 M direct generation pairs and 300 k visual‑refinement tasks. It introduces a coarse‑to‑fine Visual Reinforcement Learning (ViRL) approach that uses visual similarity across multi‑scale patches to provide an implementation‑agnostic reward, improving alignment between rendered outputs and input visuals. Experiments on diverse benchmarks show VinciCoder outperforms existing methods, and ablation studies confirm the effectiveness of ViRL.

By Xuanle Zhao, Deyang Jiang, Zhixiong Zeng, Lei Chen, Haoyue Yang, Haibo Qiu, Jing Huang, Yufeng Zhong, Liming Zheng, Yilin Cao, Lin Ma
reinforcement-learningfine-tuningmultimodalbenchmarkssafety
More like this →
arXiv Computer Vision
3d ago

Scientific Graphics Program Synthesis via Dual Self-Consistency Reinforcement Learning

arXiv:2604.06079v2 Announce Type: replace Abstract: Graphics Program Synthesis is pivotal for interpreting and editing visual data, effectively facilitating the reverse-engineering of static visuals...

By Juekai Lin, Yun Zhu, Honglin Lin, Sijing Li, Tianwei Lin, Zheng Liu, Xiaoyang Wang, Wenqiao Zhang, Lijun Wu
llmsreinforcement-learningmultimodalbenchmarkssafetydata-engineering
More like this →
arXiv Computer Vision
6d ago

RubricRM: Generative Reward Modeling via Dynamic Rubrics for Image Generation and Editing

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.

By Zijian Kan, Wei Wang, Long Luo, Bing Zhao, Xuan Ren, Weixu Qiao, Wenbo Li, Hu Wei, Lin Qu
diffusionreinforcement-learningfine-tuningbenchmarkssafety
More like this →