← Back to all news
arXiv Machine Learning By Zhenghao Zhao, Gaowen Liu, Zhiling Lan, Yan Yan

Fisher-Guided Submodular Data Selection for Continual Pre-Training of Large Language Models

Read the original on arXiv Machine Learning →

The paper introduces a Fisher-guided submodular data selection method for continual pre‑training of large language models, addressing the forgetting problem that arises when a target‑domain corpus overwrites pretrained knowledge. By decomposing candidate gradients into anchor and frontier components and optimizing a log‑determinant submodular objective, the method selects data that both improves target‑domain performance and limits forgetting. Experiments on TinyLlama‑1.1B and Llama‑3.1‑8B show that the selector achieves superior adaptation and forgetting control while using ten times fewer tokens than traditional replay strategies.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
Sep 28

Estimating and Orthogonalizing Unknown Pre-training Gradients for Continual Fine-tuning of Large Language Models

The paper introduces EoupCT, a framework that estimates and orthogonalizes unknown pre‑training gradients to mitigate catastrophic forgetting during continual fine‑tuning of large language models. It generates pseudo data most susceptible to forgetting using a learnable soft prompt with Gumbel‑Softmax, then jointly optimizes model parameters and the prompt via a first‑order Pareto optimizer to enforce orthogonality between new task updates and the estimated gradients. Experiments on multiple LLMs show that EoupCT preserves both task‑specific performance and the models’ inherent general‑purpose knowledge.

By Bing Wang, Changchun Li, Xin-Qiang Cai, Lin Yuanbo Wu, Ximing Li, Gang Niu, Masashi Sugiyama
llmsfine-tuning
More like this →
arXiv Statistics ML
Oct 2

ChainLoRA: Geometry-Preserving Task Vector Merging for Continual Learning in LLMs

ChainLoRA is a replay‑free continual learning framework for large language models that merges task vectors while preserving geometry. It uses chain‑updated training with a one‑sided orthogonality proxy to keep historical state and regularization overhead constant, and applies post‑stream adaptive SVD merging with Procrustes adaptation to separate shared and task‑specific components. Experiments demonstrate state‑of‑the‑art performance on Large and SuperNI benchmarks and competitive results on Standard CL, approaching the scores of replay‑based methods.

By Hang Yin, Haozhe Wang, Yuhua Luo, Zhangqi Pan, Xiaoxing Wang, Junchi Yan
llmsfine-tuningbenchmarks
More like this →