pith. sign in

arxiv: 2404.00213 · v2 · pith:3VUMUSSAnew · submitted 2024-03-30 · 💻 cs.CL

Injecting New Knowledge into Large Language Models via Supervised Fine-Tuning

Nick Mecklenburg , Yiyou Lin , Xiaoxiao Li , Daniel Holstein , Leonardo Nunes , Sara Malvar , Bruno Silva , Ranveer Chandra
show 4 more authors
Vijay Aski Pavan Kumar Reddy Yannam Tolga Aktas Todd Hendry
This is my paper
classification 💻 cs.CL
keywords knowledgellmsmodelsscalingacrosscoveragedatasetdomain
0
0 comments X
read the original abstract

In recent years, Large Language Models (LLMs) have shown remarkable performance in generating human-like text, proving to be a valuable asset across various applications. However, adapting these models to incorporate new, out-of-domain knowledge remains a challenge, particularly for facts and events that occur after the model's knowledge cutoff date. This paper investigates the effectiveness of Supervised Fine-Tuning (SFT) as a method for knowledge injection in LLMs, specifically focusing on the domain of recent sporting events. We compare different dataset generation strategies -- token-based and fact-based scaling -- to create training data that helps the model learn new information. Our experiments on GPT-4 demonstrate that while token-based scaling can lead to improvements in Q&A accuracy, it may not provide uniform coverage of new knowledge. Fact-based scaling, on the other hand, offers a more systematic approach to ensure even coverage across all facts. We present a novel dataset generation process that leads to more effective knowledge ingestion through SFT, and our results show considerable performance improvements in Q&A tasks related to out-of-domain knowledge. This study contributes to the understanding of domain adaptation for LLMs and highlights the potential of SFT in enhancing the factuality of LLM responses in specific knowledge domains.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Crafting Reversible SFT Behaviors in Large Language Models

    cs.LG 2026-05 unverdicted novelty 8.0

    LCDD creates sparse carriers for SFT behaviors that SFT-Eraser can reverse, with ablations showing the sparse structure enables causal control.

  2. JUDO: A Juxtaposed Domain-Oriented Multimodal Reasoner for Industrial Anomaly QA

    cs.CV 2026-05 unverdicted novelty 6.0

    JUDO enhances large multimodal models for industrial anomaly QA by juxtaposing query images with normal ones for visual comparison and using SFT plus GRPO with tailored rewards to inject domain knowledge, outperformin...

  3. SFT-then-RL Outperforms Mixed-Policy Methods for LLM Reasoning

    cs.LG 2026-04 conditional novelty 6.0

    Correcting DeepSpeed optimizer and OpenRLHF loss bugs reveals SFT-then-RL outperforms mixed-policy methods by 3.8-22.2 points on math benchmarks.

  4. Fine-Tuning Without Forgetting via Loss-Adaptive Learning Rates

    cs.LG 2026-05 unverdicted novelty 5.0

    FINCH is a loss-adaptive learning-rate schedule that reduces forgetting by 93% on average during LLM fine-tuning while matching standard task performance across several benchmarks.

  5. BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards

    cs.AI 2026-06 unverdicted novelty 4.0

    BV-Blend blends prompt-local and semantic-cluster historical reward statistics via SEM-derived weights to stabilize critic-free RL advantage estimation.

  6. Towards EnergyGPT: A Large Language Model Specialized for the Energy Sector

    cs.CL 2025-09 unverdicted novelty 3.0

    Fine-tuned LLaMA 3.1-8B variants for the energy sector outperform the base model on domain QA benchmarks, with LoRA delivering similar gains at lower training cost.