AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Generative Active Learning for Long-tail Trajectory Prediction via Controllable Diffusion Model

Daehee Park, Monu Surana, Pranav Desai, Ashish Mehta, Reuben MV John, Kuk-Jin Yoon

ICCV 2025 · Probabilistic and scientific modeling › Molecules and proteins · approach: Multimodal trajectory prediction · 4 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 real guidance (ablation)GALTraj (QCNet) · WOMDFPR5: 0.34 -> 0.22Improved
add real guidance (ablation)GALTraj (QCNet) · WOMDVaR999: 7.56 -> 6.05Improved
add gradient guidance (ablation)GALTraj (QCNet) · WOMDminFDE6: 0.601 -> 0.558Improved
add gradient guidance (ablation)GALTraj (QCNet) · WOMDFPR5: 0.26 -> 0.22Improved
add sampling weight decay and random time shift to real+gradient guidance (ablation)GALTraj (QCNet) · WOMDFPR5: 0.28 -> 0.22Improved
add sampling weight decay and random time shift to real+gradient guidance (ablation)GALTraj (QCNet) · WOMDminFDE6: 0.604 -> 0.558Improved
add unguided generative augmentation (naive concatenation) (ablation)QCNet vanilla · WOMDFPR5: 0.42 -> 0.38Improved
replace naive generation with tail-aware generation (ablation)QCNet · WOMDTop 1% minADE6: 4.56 -> 3.43Improved
resampling tail samples instead of vanilla trainingQCNet · WOMDminFDE6: 0.654 -> 0.668Hurt
GALTraj training instead of vanillaMTR · WOMDFPR5: 0.99 -> 0.65Improved
GALTraj training instead of vanillaQCNet · AV2FPR5: 0.35 -> 0.19Improved
GALTraj training instead of vanilla on training splitQCNet · WOMD train splitTop 1%: 7.38 -> 2.29Improved

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