AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

FACET: A Fragment-Aware Conformer Ensemble Transformer

Duy Minh Ho Nguyen, Trung Quoc Nguyen, Ha Thi Hong Le, Mai Thanh Nhat Truong, TrungTin Nguyen, Nhat Ho, Khoa D Doan, Duy Duong-Tran, Li Shen, Daniel Sonntag, James Zou, Mathias Niepert and 2 more

ICLR 2026 · Graph learning › Graph self-supervised learning · approach: Molecular graph pretraining

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 fragment graph (ablation)FACET · ESOLMSE: 0.531 -> 0.516Improved
add fragment graph (ablation)FACET · FreeSolvMSE: 1.072 -> 0.967Improved
add fragment graph (ablation)FACET · KrakenMSE: 0.247 -> 0.238Improved
add fragment graph inside graph transformer (ablation)FACET · FreeSolvMSE: 0.973 -> 0.967No clear effect
add adaptors aligning 3D and 2D features (ablation)FACET · ESOLMSE: 0.546 -> 0.516Improved
add adaptors aligning 3D and 2D features (ablation)FACET · FreeSolvMSE: 1.085 -> 0.967Improved
add adaptors aligning 3D and 2D features (ablation)FACET · KrakenMSE: 0.262 -> 0.238Improved
train in separate stages instead of merging all steps (ablation)FACET · FreeSolvMSE: 1.26 -> 0.97Improved
add FGW supervision of graph transformer (ablation)FACET · ESOLMSE: 0.54 -> 0.52Improved
keep steps 2 and 3 separate rather than merged (ablation)FACET · FreeSolvMSE: 0.87 -> 0.97Hurt
use FACET instead of CONAN-FGW (SchNet)FACET · FreeSolvMSE: 1.068 -> 0.967Improved

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