You Only Merge Once: Learning the Pareto Set of Preference-Aware Model Merging

• URL: http://arxiv.org/abs/2408.12105v1
• Date: Thu, 22 Aug 2024 03:41:14 GMT
• Title: You Only Merge Once: Learning the Pareto Set of Preference-Aware Model Merging
• Authors: Weiyu Chen, James Kwok,
• Abstract summary: We propose preference-aware model merging in which the performance of the merged model on each base model's task is treated as an objective. In only one merging process, the proposed parameter-efficient structure can generate the whole set of merged models. We show that the proposed preference-aware model merging can obtain a diverse set of trade-off models and outperforms state-of-the-art model merging baselines.
• Score: 11.186194228460273
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Model merging, which combines multiple models into a single model, has gained increasing popularity in recent years. By efficiently integrating the capabilities of various models without their original training data, this significantly reduces the parameter count and memory usage. However, current methods can only produce one single merged model. This necessitates a performance trade-off due to conflicts among the various models, and the resultant one-size-fits-all model may not align with the preferences of different users who may prioritize certain models over others. To address this issue, we propose preference-aware model merging, and formulate this as a multi-objective optimization problem in which the performance of the merged model on each base model's task is treated as an objective. In only one merging process, the proposed parameter-efficient structure can generate the whole Pareto set of merged models, each representing the Pareto-optimal model for a given user-specified preference. Merged models can also be selected from the learned Pareto set that are tailored to different user preferences. Experimental results on a number of benchmark datasets demonstrate that the proposed preference-aware Pareto Merging can obtain a diverse set of trade-off models and outperforms state-of-the-art model merging baselines.

Related papers

• Exploring Model Kinship for Merging Large Language Models [52.01652098827454]
  We introduce model kinship, the degree of similarity or relatedness between Large Language Models. We find that there is a certain relationship between model kinship and the performance gains after model merging. We propose a new model merging strategy: Top-k Greedy Merging with Model Kinship, which can yield better performance on benchmark datasets.
  arXiv  Detail & Related papers  (2024-10-16T14:29:29Z)
• What Matters for Model Merging at Scale? [94.26607564817786]
  Model merging aims to combine multiple expert models into a more capable single model. Previous studies have primarily focused on merging a few small models. This study systematically evaluates the utility of model merging at scale.
  arXiv  Detail & Related papers  (2024-10-04T17:17:19Z)
• Parameter Competition Balancing for Model Merging [13.66727853299506]
  PCB-Merging is a training-free technique that adjusts the coefficients of each parameter for effective model merging. PCB-Merging achieves substantial performance enhancements across multiple modalities, domains, model sizes, number of tasks, fine-tuning forms, and large language models.
  arXiv  Detail & Related papers  (2024-10-03T11:17:58Z)
• PLeaS -- Merging Models with Permutations and Least Squares [43.17620198572947]
  We propose a new two-step algorithm to merge models-termed PLeaS. PLeaS partially matches nodes in each layer by maximizing alignment. It computes the weights of the merged model as a layer-wise Least Squares solution.
  arXiv  Detail & Related papers  (2024-07-02T17:24:04Z)
• EMR-Merging: Tuning-Free High-Performance Model Merging [55.03509900949149]
  We show that Elect, Mask & Rescale-Merging (EMR-Merging) shows outstanding performance compared to existing merging methods. EMR-Merging is tuning-free, thus requiring no data availability or any additional training while showing impressive performance.
  arXiv  Detail & Related papers  (2024-05-23T05:25:45Z)
• Continuous Language Model Interpolation for Dynamic and Controllable Text Generation [7.535219325248997]
  We focus on the challenging case where the model must dynamically adapt to diverse -- and often changing -- user preferences. We leverage adaptation methods based on linear weight, casting them as continuous multi-domain interpolators. We show that varying the weights yields predictable and consistent change in the model outputs.
  arXiv  Detail & Related papers  (2024-04-10T15:55:07Z)
• Dataless Knowledge Fusion by Merging Weights of Language Models [51.8162883997512]
  Fine-tuning pre-trained language models has become the prevalent paradigm for building downstream NLP models. This creates a barrier to fusing knowledge across individual models to yield a better single model. We propose a dataless knowledge fusion method that merges models in their parameter space.
  arXiv  Detail & Related papers  (2022-12-19T20:46:43Z)
• Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers [66.36045164286854]
  We analyze a set of existing bias features and demonstrate there is no single model that works best for all the cases. By choosing an appropriate bias model, we can obtain a better robustness result than baselines with a more sophisticated model design.
  arXiv  Detail & Related papers  (2022-10-28T17:52:10Z)
• When Ensembling Smaller Models is More Efficient than Single Large Models [52.38997176317532]
  We show that ensembles can outperform single models with both higher accuracy and requiring fewer total FLOPs to compute. This presents an interesting observation that output diversity in ensembling can often be more efficient than training larger models.
  arXiv  Detail & Related papers  (2020-05-01T18:56:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.