AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization

qing liu, Xinrui Chen, Weiyao Zhu, Yi Du, Ou Wu

NeurIPS 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Low-rank and model-merging adaptation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes DRIVE, a data-weighting fine-tuning method that uses contribution- and diversity-aware sample weights with prior regularization to reduce catastrophic forgetting when source pre-training data is inaccessible.

Problem

Single-stage fine-tuning causes catastrophic forgetting of pre-trained knowledge when the source pre-training data is unavailable; existing data weighting methods lack a principled way to trade off knowledge retention against new-task adaptation.

Approach

Weights fine-tuning samples by their contribution and diversity and adds a prior regularization term, giving a unified differentiable objective that can accommodate auxiliary signals; the abstract indicates this explicitly arbitrates retention versus adaptation.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove task-adaptation component (lambda=1) (ablation)DRIVE · MetaMathQA, Llama-3.1-8Btarget accuracy: 77.03 -> 73.19Improved
remove anti-forgetting component (lambda=0) (ablation)DRIVE · MetaMathQA, Llama-3.1-8Bgeneral capability: 57.51 -> 52.24Improved
remove diversity regularization (eta=0) (ablation)DRIVE · MetaMathQA, Llama-3.1-8Boverall accuracy: 67.27 -> 65.48Improved
remove uniform KL prior regularization (ablation)DRIVE · MetaMathQA, Llama-3.1-8Boverall accuracy: 67.27 -> 62.29Improved
replace differentiable diversity regularizer with discrete clustering (DRIVE-Cluster) (ablation)DRIVE · MetaMathQA, Llama-3.1-8Boverall accuracyImproved
use entropy-informed prior instead of uniform prior (ablation)DRIVE · MetaMathQA, Llama-3.1-8Bgeneral/overall accuracyNo clear effect
decay lambda from 1 to 0 instead of best static lambda=0.8 (ablation)DRIVE · ResNet-50 visionaverage accuracyImproved
increase eta beyond 1.0 (ablation)DRIVE · hyperparameter sensitivitygeneralizationHurt
use gradient-norm contribution scores instead of loss (ablation)DRIVE · Llama-3.2-3Boverall accuracy: 57.69 -> 57.99 at 2.12x costMixed
use entropy-based contribution scores instead of loss (ablation)DRIVE · Llama-3.2-3Boverall accuracy: 57.69 -> 55.14Hurt
shuffle contribution score ranking (ablation)DRIVE · MetaMathQA, Llama-3.1-8Boverall accuracy: 67.29 -> 65.12Hurt
reverse contribution score ranking (ablation)DRIVE · MetaMathQA, Llama-3.1-8Boverall accuracy: 67.29 -> 60.23Hurt

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