Error-Subspace Calibration for Multi-Task Model Merging

Model merging combines task-specific expert models into a single model without jointly retraining it on their original datasets, but often incurs a performance gap relative to the individual experts. Post-merging calibration can reduce this gap through task-specific representation alignment, but the alignment objective alone does not specify which feature directions to correct. We propose *Error-Subspace Calibration* (**ESCal**), a post-merging calibration framework that identifies low-dimensional error subspaces from weight differences between expert and merged models. At each intermediate block boundary, a **Bridge** module projects the merged representation onto the next block's input error subspace and maps these coordinates to an additive correction in the current block's output error subspace. The subspace bases remain fixed, so calibration learns only these linear maps, yielding a parameter-efficient mechanism for feature correction. We find that **ESCal** brings the merged model closer to the experts in both representations and predictions, indicating more faithful recovery of task-specific behavior. Experiments on vision and natural language understanding benchmarks show that **ESCal** improves merged-model performance across diverse architectures and merging methods using a small calibration set for each task.