CIP training reduces shortcut reliance in TCR-pMHC models by 39.7% on family-held-out tests via invariance to non-anchor edits and sensitivity to anchor disruptions, reaching AUROC 0.831 and CFC 0.724.
Modeling tcr-pmhc binding with dual encoders and cross-attention fusion,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A dual-encoder TCR-pMHC model with temperature scaling and conformal abstention achieves AUROC 0.813, ECE 0.043, and reduces error from 18.7% to 10.9% at 80% coverage under epitope-held-out evaluation.
citing papers explorer
-
Counterfactual Peptide Editing for Causal TCR--pMHC Binding Inference
CIP training reduces shortcut reliance in TCR-pMHC models by 39.7% on family-held-out tests via invariance to non-anchor edits and sensitivity to anchor disruptions, reaching AUROC 0.831 and CFC 0.724.
-
Calibrated Abstention for Reliable TCR--pMHC Binding Prediction under Epitope Shift
A dual-encoder TCR-pMHC model with temperature scaling and conformal abstention achieves AUROC 0.813, ECE 0.043, and reduces error from 18.7% to 10.9% at 80% coverage under epitope-held-out evaluation.