AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Prompt2Effect: Training-Free LoRA Synthesis for Controllable Video Effects

Xiaomeng Yang, Yanyu Li, Gordon Qian, Ivan Skorokhodov, Viacheslav Ivanov, Avalon Vinella, Xuan Zhang, Yanzhi Wang, Sergey Tulyakov, Anil Kag

ECCV 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: LoRA variants and theory

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 weight-driven base-weight token input (vs noise embeddings) (ablation)Prompt2Effect · IID effectsVLM Overall: 46.14 -> 56.93Improved
add weight-driven base-weight token input (vs noise embeddings) (ablation)Prompt2Effect · OOD effectsVLM Overall: 43.58 -> 48.87Improved
use one joint hypernetwork instead of per-block ensemble (ablation)Prompt2Effect · IID effectsVLM Overall: 51.29 -> 56.93Improved
use one joint hypernetwork instead of per-block ensemble (ablation)Prompt2Effect · OOD effectsVLM Overall: 45.11 -> 48.87Improved
predict SVD-canonicalized LoRA factors instead of raw LoRA (ablation)Prompt2Effect · training NMSENMSE convergenceImproved
compress weight tokens to n=256 instead of n=128 (ablation)Prompt2Effect · IID effectsVLM Overall: 49.75 -> 56.93Improved
compress weight tokens to n=256 instead of n=512 (ablation)Prompt2Effect · IID effectsVLM Overall: 56.22 -> 56.93No clear effect
increase tokens from n=256 to n=512 (ablation)Prompt2Effect · OOD effectsAesthetic Quality: 53.44 -> 54.02Mixed
use predicted weights as init, 100 steps, vs LoRA from scratch 100 stepsPrompt2Effect · OOD effectsVLM Overall: 52.28 -> 56.33Improved
use predicted weights as init, 50 steps, vs LoRA from scratch 100 stepsPrompt2Effect · OOD effectsVLM Overall: 52.28 -> 54.89Improved
fast adaptation Init-100 vs 1000-step LoRAPrompt2Effect · OOD effectsCLIP Score: 25.73 -> 25.82Improved
zero-shot Prompt2Effect vs base I2V modelPrompt2Effect · IID effectsVLM Overall: 42.57 -> 56.93Improved

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