paper-with-me

Papers

Preserving Knowledge in Large Language Model with Model-Agnostic Self-Decompression

2024-06-17 · Zilun Zhang, Yutao Sun, Tiancheng Zhao, Leigang Sha, Ruochen Xu, Kyusong Lee, Jianwei Yin

Humans can retain old knowledge while learning new information, but Large Language Models (LLMs) often suffer from catastrophic forgetting when post-pretrained or supervised fine-tuned (SFT) on domain-specific data. Moreover, for Multimodal Large Language Models (MLLMs) which are composed of the LLM base and visual projector (e.g. LLaVA), a significant decline in performance on language benchmarks was observed compared to their single-modality counterparts. To address these challenges, we introduce a novel model-agnostic self-decompression method, Tree Generation (TG), that decompresses knowledge within LLMs into the training corpus. This paper focuses on TG-SFT, which can synthetically generate SFT data for the instruction tuning steps. By incorporating the dumped corpus during SFT for MLLMs, we significantly reduce the forgetting problem.

📄 PDF Abstract BibTeX arXiv:2406.11354

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLarge Language Modelmodel

Methods 이 논문이 사용한 방법론

BASE 설명 없음
SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…

Similar Papers 제목 키워드 기반

How to Train Private Clinical Language Models: A Comparative Study of Privacy-Preserving Pipelines for ICD-9 Coding

2025-11-18 · Mathieu Dufour, Andrew Duncan arxiv

Large language models trained on clinical text risk exposing sensitive patient information, yet differential privacy (DP) methods often severely degrade the diagnostic accuracy needed for deployment. Despite rapid progre…

Knowledge DistillationText Generation

LLMs can Compress LLMs: Adaptive Pruning by Agents

2026-01-14 · Sai Varun Kodathala, Rakesh Vunnam arxiv

As Large Language Models (LLMs) continue to scale, post-training pruning has emerged as a promising approach to reduce computational costs while preserving performance. Existing methods such as SparseGPT and Wanda achiev…

One Mind, Many Tongues: A Deep Dive into Language-Agnostic Knowledge Neurons in Large Language Models

2024-11-26 · Pengfei Cao, YuHeng Chen, Zhuoran Jin, Yubo Chen 외

Large language models (LLMs) have learned vast amounts of factual knowledge through self-supervised pre-training on large-scale corpora. Meanwhile, LLMs have also demonstrated excellent multilingual capabilities, which c…

knowledge editing

Quantifying Self-diagnostic Atomic Knowledge in Chinese Medical Foundation Model: A Computational Analysis

2023-10-18 · Yaxin Fan, Feng Jiang, Benyou Wang, Peifeng Li 외

Foundation Models (FMs) have the potential to revolutionize the way users self-diagnose through search engines by offering direct and efficient suggestions. Recent studies primarily focused on the quality of FMs evaluate…

DiagnosticInstruction Following

Self-Tuning: Instructing LLMs to Effectively Acquire New Knowledge through Self-Teaching

2024-06-10 · Xiaoying Zhang, Baolin Peng, Ye Tian, Jingyan Zhou 외

Large language models (LLMs) often struggle to provide up-to-date information due to their one-time training and the constantly evolving nature of the world. To keep LLMs current, existing approaches typically involve co…

Memorization