Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation
Authors:
Preprint
Abstract
Large Language Models (LLMs) face the "knowledge cutoff" challenge, where their frozen parametric memory prevents direct internalization of new information. While Supervised Fine-Tuning (SFT) is commonly used to update model knowledge, it often updates factual content without reliably improving the model's ability to use the newly incorporated information for question answering or decision-making. Reinforcement Learning (RL) is essential for acquiring reasoning skills; however, its high computational cost makes it impractical for efficient online adaptation. We empirically observe that the parameter updates induced by SFT and RL are nearly orthogonal. Based on this observation, we propose Parametric Skill Transfer (PaST), a framework that supports modular skill transfer for efficient and effective knowledge adaptation. By extracting a domain-agnostic Skill Vector from a source domain, we can linearly inject knowledge manipulation skills into a target model after it has undergone lightweight SFT on new data. Experiments on knowledge-incorporation QA (SQuAD, LooGLE) and agentic tool-use benchmarks (ToolBench) demonstrate the effectiveness of our method. On SQuAD, PaST outperforms the state-of-the-art self-editing SFT baseline by up to 9.9 points. PaST further scales to long-context QA on LooGLE with an 8.0-point absolute accuracy gain, and improves zero-shot ToolBench success rates by +10.3 points on average with consistent gains across tool categories, indicating strong scalability and cross-domain transferability of the Skill Vector.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models2024
- Toolformer: Language Models Can Teach Themselves to Use Tools2023
- Editing Large Language Models: Problems, Methods, and Opportunities2023
- Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy2023
- API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs2023
- Test-Time Training with Masked Autoencoders2022
- How Much Knowledge Can You Pack Into the Parameters of a Language Model?2020
- SQuAD: 100,000+ Questions for Machine Comprehension of Text2016