{
  "annotation_id": "4cbc0361-978c-45c3-a07e-45e7069ccf66",
  "date_created": "2026-02-17T05:53:08.878000Z",
  "date_modified": "2026-02-17T05:53:08.878000Z",
  "file_hash": "23ed6c51e2ef9e787efcf1fb6440e28c7ea8dcd2d42f14cbdca4b52a06a3b07c",
  "private": false,
  "record": {
    "abstract": "Multi-hop question answering is widely used to evaluate the reasoning capabilities of large language models (LLMs), as it requires integrating multiple pieces of supporting knowledge to arrive at a correct answer. While prior work has explored different mechanisms for providing knowledge to LLMs, such as finetuning and retrieval-augmented generation (RAG), their relative effectiveness for multi-hop question answering remains insufficiently understood, particularly when the required knowledge is temporally novel.\n  In this paper, we systematically compare parametric and non-parametric knowledge injection methods for open-domain multi-hop question answering. We evaluate unsupervised fine-tuning (continual pretraining), supervised fine-tuning, and retrieval-augmented generation across three 7B-parameter open-source LLMs. Experiments are conducted on two benchmarks: QASC, a standard multi-hop science question answering dataset, and a newly constructed dataset of over 10,000 multi-hop questions derived from Wikipedia events in 2024, designed to test knowledge beyond the models\u0027 pretraining cutoff.\n  Our results show that unsupervised fine-tuning provides only limited gains over base models, suggesting that continual pretraining alone is insufficient for improving multi-hop reasoning accuracy. In contrast, retrieval-augmented generation yields substantial and consistent improvements, particularly when answering questions that rely on temporally novel information. Supervised fine-tuning achieves the highest overall accuracy across models and datasets. These findings highlight fundamental differences in how knowledge injection mechanisms support multi-hop question answering and underscore the importance of retrieval-based methods when external or compositional knowledge is required.",
    "arxiv_id": "2601.07054",
    "authors": [
      "Zhuoyi Yang",
      "Yurun Song",
      "Iftekhar Ahmed",
      "Ian Harris"
    ],
    "categories": [
      "cs.CL",
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "title": "Fine-Tuning vs. RAG for Multi-Hop Question Answering with Novel Knowledge",
    "url": "https://arxiv.org/abs/2601.07054",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "d883dae2-10f2-428d-8684-819122a33137",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}