AI Paper Atlas

Research map / Video / Video generation

LottieGPT: Tokenizing Vector Animation for Autoregressive Generation

Junhao Chen, Kejun Gao, Yuehan Cui, Mingze Sun, Mingjin Chen, Shaohui Wang, Xiaoxiao Long, Fei Ma, Qi Tian, Hao Zhao, Ruqi Huang

CVPR 2026 · Video › Video generation · approach: Autoregressive long video generation · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the Lottie tokenizer instead of raw Lottie JSON text (ablation)Qwen2.5-VL-7B · LottieBench animation, text-onlyvalid rate (%): 22.61% -> 78.35%Improved
add the Lottie tokenizer instead of raw Lottie JSON text (ablation)Qwen2.5-VL-7B · LottieBench animation, text-onlyCLIP: 0.7456 -> 0.8156Improved
add the Lottie tokenizer instead of raw Lottie JSON text (ablation)Qwen2.5-VL-7B · LottieBench animation, text-onlyDINOv2: 0.4789 -> 0.7715Improved
add Stage 2 animation training after static Stage 1 (ablation)LottieGPT-7B · LottieBench animation, text-onlyvalid rate (%): 78.35% -> 96.96%Improved
add Stage 2 animation training after static Stage 1 (ablation)LottieGPT-7B · LottieBench animation, text-onlyLPIPS: 0.1925 -> 0.0366Improved
add Stage 2 animation training after static Stage 1 (ablation)LottieGPT-7B · LottieBench animation, text-onlyJSON structure: 0.2347 -> 0.8062Improved
add the Lottie tokenizer instead of QwenVL tokenizer on Lottie JSONtokenizer · MMSVG-iconavg tokens (k): 1.6 k -> 1.3 kImproved
add numeric quantization to the Lottie tokenizerLottie tokenizer · LottieAnimationavg tokens (k): 17.4 k -> 6.6 kImproved
use Lottie JSON instead of SVG code as training textQwenVL tokenizer · MMSVG-iconavg tokens (k): 2.6 k -> 1.6 kImproved
simplify Lottie JSON (prune fields, round values)Lottie dataset · LottieAnimationavg JSON size: 60.72 KB -> 39.87 KBImproved
joint training on mixed static and dynamic data instead of two stages (ablation)LottieGPT · training stabilityconvergenceHurt
train on Lottie JSON rather than SVG-based SVG generator (LottieGPT vs OmniSVG)LottieGPT-7B-Stage1 · LottieBench static, text-onlyCLIP: 0.8321 -> 0.9331Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy