כתבה
arXiv cs.CL ·
LadderEdit: דחיסת עריכה יעילה בזיכרון
LadderEdit: Edit-Level Residual Compression for Memory-Efficient Lifelong Editing of LLMs
LadderEdit הוא שיטה לדחיסת עריכות במודלים LLM. היא מאפשרת אחסון עריכות רבות בזיכרון נמוך. השיטה נבדקה על מודלים LLaMA-3-8B, Mistral-7B ו-Qwen2.5-7B.
תקציר מקורי באנגליתarXiv:2610.11160v1 Announce Type: new Abstract: Lifelong editing of LLMs requires storing thousands of edits after acquisition. A widely used family of approaches attaches one LoRA adapter per edit, which preserves behavior but grows linearly in storage. To address this challenge, we propose LadderEdit, a method that compresses each LoRA adapter after it is acquired. Each edit is first stored at low rank as a cheap sketch. We then check whether this sketch still satisfies the rewrite, generalization, and locality contract on probe prompts. Edits that pass keep the sketch; those that fail are promoted to a higher rank along a ladder until the contract is met. Because every edit retains some representation, coverage is maintained, and only hard edits consume more rank. Across ZsRE, CounterFa
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית