Data-Aware Model Merging in the Essential Subspace for Multi-Task Knowledge Aggregation

arXiv:2606.19164 · cs.LG, cs.AI · Submitted 2026-06-17 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-06-17

Updated: 2026-10-13

Code: https://github.com/kiddo127/ESM

Terminology

• Model Merging — 2× in this paper · explained in 2 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Core — 1× in this paper · explained in 2 episodes
• Essential Subspace Decomposition — 1× in this paper
• Essential Subspace Merging — 1× in this paper

Sources

• Energy-Structured Low-Rank Adaptation for Continual Learning
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Deep Learning for Classical Japanese Literature
• RoBERTa: A Robustly Optimized BERT Pretraining Approach
• The Llama 3 Herd of Models
• Pointer Sentinel Mixture Models
• Instruction-Following Evaluation for Large Language Models
• Training Verifiers to Solve Math Word Problems
• Evaluating Large Language Models Trained on Code
• Program Synthesis with Large Language Models
• HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks