Elamur Research⌕
☰
home / papers / 2024 / Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs

Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs

Paper · 2024 · Oded Ovadia, Menachem Brief, Moshik Mishaeli et al.

Summary

The paper asks how best to add factual knowledge to a pre-trained LLM when a relevant text corpus is available. Two strategies are compared: unsupervised fine-tuning (continued next-token training on the corpus) and retrieval-augmented generation (RAG), where relevant passages are inserted into the prompt at query time. Knowledge is measured as accuracy on multiple-choice questions. A model counts as knowledgeable when it clearly beats random guessing. The authors test three open models, among them Llama2 and Orca2. They use two kinds of material: MMLU subject areas, which the models likely saw during pre-training, and a newly built current-events question set about facts the models had never seen. All knowledge bases come from Wikipedia. RAG wins consistently in both settings. Fine-tuning gives modest gains at best, and on the new facts it can even hurt. Augmenting training data with paraphrases steadily improves fine-tuning. This suggests that new facts must appear in many phrasings before a model can absorb them. For practitioners, the takeaway is that retrieval is the more dependable route for domain adaptation and keeping models current.

Key points

  • Knowledge is defined as plain accuracy on multiple-choice questions. With four answer options, the random baseline is 0.25.
  • RAG outperformed unsupervised fine-tuning on both the MMLU-based tasks and the current-events task. Combining RAG with fine-tuning was worse than RAG alone.
  • On the current-events task, standard fine-tuning failed to improve Llama2 and actually lowered its accuracy noticeably.
  • When training data was augmented with paraphrases, accuracy rose monotonically with the number of paraphrases for every model tested. This supports the hypothesis that repeating a fact in varied forms is needed for learning it.
  • Limitations include results that depend on hyperparameters, only three models tested, and all source texts drawn from Wikipedia. The strong showing of Orca2 compared with base Llama2 hints that extra alignment may help fine-tuned models.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 9,086 worksArtificial Intelligence · 8,889 worksArts and Humanities · 9,296 worksLanguage and Linguistics · 9,206 worksComputer Vision and Pattern Recognition · 1,282 worksInformation Systems · 523 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 1,343 worksInformation retrieval & search · 145 works

Cite

Cite as Oded Ovadia, Menachem Brief, Moshik Mishaeli and Oren Elisha. 2024. Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2024.emnlp-main.15

@inproceedings{ovadia2024fine,
  title = {{Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs}},
  author = {Oded Ovadia, Menachem Brief, Moshik Mishaeli and Oren Elisha},
  booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing},
  year = {2024},
  doi = {10.18653/v1/2024.emnlp-main.15},
  url = {https://doi.org/10.18653/v1/2024.emnlp-main.15}
}

References

54 references, 25 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Giusepppe Attardi. 2015. Wikiextractor. https://github.com/attardi/wikiextractor. ↗
  2. Leo Gao, Jonathan Tow, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Kyle McDonell, Niklas Muennighoff, Jason Phang, Laria Reynolds, Eric Tang, Anish Thite, Ben Wang, Kevin Wang, and Andy Zou. 2021. A framework for few-shot language model evaluation. find
  3. Ian J Goodfellow, Mehdi Mirza, Da Xiao, Aaron Courville, and Yoshua Bengio. 2013. An empirical investigation of catastrophic forgetting in gradient-based neural networks. arXiv preprint arXiv:1312.6211. find
  4. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR).
  5. Linmei Hu, Zeyi Liu, Ziwang Zhao, Lei Hou, Liqiang Nie, and Juanzi Li. 2023. A survey of knowledge enhanced pre-trained language models. IEEE Transactions on Knowledge and Data Engineering. find
  6. Lukas Berglund, Meg Tong, Max Kaufmann, Mikita Balesni, Asa Cooper Stickland, Tomasz Korbak, and Owain Evans. 2023. The reversal curse: Llms trained on" a is b" fail to learn" b is a". arXiv preprint arXiv:2309.12288. find
  7. Quzhe Huang, Mingxu Tao, Zhenwei An, Chen Zhang, Cong Jiang, Zhibin Chen, Zirui Wu, and Yansong Feng. 2023. Lawyer llama technical report. arXiv preprint arXiv:2305.15062.
  8. Sanyuan Chen, Yutai Hou, Yiming Cui, Wanxiang Che, Ting Liu, and Xiangzhan Yu. 2020. Recall and learn: Fine-tuning deep pretrained language models with less forgetting. arXiv preprint arXiv:2004.12651. find
  9. Xiang Chen, Ningyu Zhang, Xin Xie, Shumin Deng, Yunzhi Yao, Chuanqi Tan, Fei Huang, Luo Si, and Huajun Chen. 2022. Knowprompt: Knowledge-aware prompt-tuning with synergistic optimization for relation extraction. In Proceedings of the ACM Web conference 2022, pages 2778–2788. find
  10. Yanda Chen, Ruiqi Zhong, Sheng Zha, George Karypis, and He He. 2021. Meta-learning via language model in-context tuning. arXiv preprint arXiv:2110.07814. find
  11. Yew Ken Chia, Pengfei Hong, Lidong Bing, and Soujanya Poria. 2023. Instructeval: Towards holistic evaluation of instruction-tuned large language models. arXiv preprint arXiv:2306.04757.
  12. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, et al. 2022. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416.
  13. Roi Cohen, Mor Geva, Jonathan Berant, and Amir Globerson. 2023. Crawling the internal knowledge-base of language models. arXiv preprint arXiv:2301.12810. find
  14. Dheeru Dua, Yizhong Wang, Pradeep Dasigi, Gabriel Stanovsky, Sameer Singh, and Matt Gardner. 2019. Drop: A reading comprehension benchmark requiring discrete reasoning over paragraphs. arXiv preprint arXiv:1903.00161.
  15. Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al. 2023. Mistral 7b. arXiv preprint arXiv:2310.06825. find
  16. Jeff Johnson, Matthijs Douze, and Hervé Jégou. 2019. Billion-scale similarity search with GPUs. IEEE Transactions on Big Data, 7(3):535–547.
  17. Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, and Colin Raffel. 2023. Large language models struggle to learn long-tail knowledge. In International Conference on Machine Learning, pages 15696–15707. PMLR.
  18. James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, et al. 2017. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of sciences, 114(13):3521… find
  19. Andrew K Lampinen, Ishita Dasgupta, Stephanie CY Chan, Kory Matthewson, Michael Henry Tessler, Antonia Creswell, James L McClelland, Jane X Wang, and Felix Hill. 2022. Can language models learn from explanations in context? arXiv preprint arXiv:2204.02329.
  20. Anne Lauscher, Olga Majewska, Leonardo FR Ribeiro, Iryna Gurevych, Nikolai Rozanov, and Goran Glavaš. 2020. Common sense or world knowledge? investigating adapter-based knowledge injection into pretrained transformers. arXiv preprint arXiv:2005.11787. find
  21. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, et al. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems, 33:9459–9474. find
  22. Weijie Liu, Peng Zhou, Zhe Zhao, Zhiruo Wang, Qi Ju, Haotang Deng, and Ping Wang. 2020. K-bert: Enabling language representation with knowledge graph. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 2901–2908.
  23. Yun Luo, Zhen Yang, Fandong Meng, Yafu Li, Jie Zhou, and Yue Zhang. 2023. An empirical study of catastrophic forgetting in large language models during continual fine-tuning. arXiv preprint arXiv:2308.08747. find
  24. Sewon Min, Mike Lewis, Luke Zettlemoyer, and Hannaneh Hajishirzi. 2021. Metaicl: Learning to learn in context. arXiv preprint arXiv:2110.15943. find
  25. Swaroop Mishra, Daniel Khashabi, Chitta Baral, and Hannaneh Hajishirzi. 2021. Cross-task generalization via natural language crowdsourcing instructions. arXiv preprint arXiv:2104.08773. find
  26. Arindam Mitra, Luciano Del Corro, Shweti Mahajan, Andres Codas, Clarisse Simoes, Sahaj Agrawal, Xuxi Chen, Anastasia Razdaibiedina, Erik Jones, Kriti Aggarwal, et al. 2023. Orca 2: Teaching small language models how to reason. arXiv preprint arXiv:2311.11045.
  27. Arvind Neelakantan, Tao Xu, Raul Puri, Alec Radford, Jesse Michael Han, Jerry Tworek, Qiming Yuan, Nikolas A. Tezak, Jong Wook Kim, Chris Hallacy, Johannes Heidecke, Pranav Shyam, Boris Power, Tyna Eloundou Nekoul, Girish Sastry, Gretchen Krueger, David P. Schnurr, Felipe Petroski Such, Kenny Sai-Ki…
  28. Ha-Thanh Nguyen. 2023. A brief report on lawgpt 1.0: A virtual legal assistant based on gpt-3. arXiv preprint arXiv:2302.05729. find
  29. Harsha Nori, Nicholas King, Scott Mayer McKinney, Dean Carignan, and Eric Horvitz. 2023. Capabilities of gpt-4 on medical challenge problems. ArXiv, abs/2303.13375.
  30. OpenAI. 2023. Gpt-4 technical report. abs/2303.08774.
  31. ArXiv, Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. 2022. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730–27744.
  32. Fabio Petroni, Tim Rocktäschel, Patrick Lewis, Anton Bakhtin, Yuxiang Wu, Alexander H Miller, and Sebastian Riedel. 2019. Language models as knowledge bases? arXiv preprint arXiv:1909.01066.
  33. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9. find
  34. Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D Manning, and Chelsea Finn. 2023. Direct preference optimization: Your language model is secretly a reward model. arXiv preprint arXiv:2305.18290.
  35. Keisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. 2021. Winogrande: An adversarial winograd schema challenge at scale. Communications of the ACM, 64(9):99–106. find
  36. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347. find
  37. Saket Sharma, Aviral Joshi, Namrata Mukhija, Yiyun Zhao, Hanoz Bhathena, Prateek Singh, Sashank Santhanam, and Pritam Biswas. 2022. Systematic review of effect of data augmentation using paraphrasing on named entity recognition. In NeurIPS 2022 Workshop on Synthetic Data for Empowering ML Research. find
  38. Connor Shorten, Taghi M. Khoshgoftaar, and Borko Furht. 2021. Text data augmentation for deep learning. Journal of Big Data, 8. find
  39. Karan Singhal, Shekoofeh Azizi, Tao Tu, S Sara Mahdavi, Jason Wei, Hyung Won Chung, Nathan Scales, Ajay Tanwani, Heather Cole-Lewis, Stephen Pfohl, et al. 2023a. Large language models encode clinical knowledge. Nature, 620(7972):172–180.
  40. Karan Singhal, Tao Tu, Juraj Gottweis, Rory Sayres, Ellery Wulczyn, Le Hou, Kevin Clark, Stephen Pfohl, Heather Cole-Lewis, Darlene Neal, et al. 2023b. Towards expert-level medical question answering with large language models. arXiv preprint arXiv:2305.09617.
  41. Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R Brown, Adam Santoro, Aditya Gupta, Adrià Garriga-Alonso, et al. 2022. Beyond the imitation game: Quantifying and extrapolating the capabilities of language models. arXiv preprint arXiv:2206.04615.
  42. Yiming Tan, Dehai Min, Yu Li, Wenbo Li, Nan Hu, Yongrui Chen, and Guilin Qi. 2023. Can chatgpt replace traditional kbqa models? an in-depth analysis of the question answering performance of the gpt llm family. In International Semantic Web Conference, pages 348–367. Springer. find
  43. Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B Hashimoto. 2023. Alpaca: A strong, replicable instruction-following model. Stanford Center for Research on Foundation Models. https://crfm.stanford. edu/2023/03/13/alpaca. html, 3(6):… ↗
  44. Kushal Tirumala, Aram H. Markosyan, Luke Zettlemoyer, and Armen Aghajanyan. 2022. Memorization without overfitting: Analyzing the training dynamics of large language models. ArXiv, abs/2205.10770.
  45. Wenhao Yu, Chenguang Zhu, Zaitang Li, Zhiting Hu, Qingyun Wang, Heng Ji, and Meng Jiang. 2022. A survey of knowledge-enhanced text generation. ACM Computing Surveys, 54(11s):1–38. find
  46. Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, et al. 2023. Lima: Less is more for alignment. arXiv preprint arXiv:2305.11206. find
  47. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288. find
  48. Lewis Tunstall, Edward Beeching, Nathan Lambert, Nazneen Rajani, Kashif Rasul, Younes Belkada, Shengyi Huang, Leandro von Werra, Clémentine Fourrier, Nathan Habib, et al. 2023. Zephyr: Direct distillation of lm alignment. arXiv preprint arXiv:2310.16944. find
  49. Cunxiang Wang, Xiaoze Liu, Yuanhao Yue, Xiangru Tang, Tianhang Zhang, Cheng Jiayang, Yunzhi Yao, Wenyang Gao, Xuming Hu, Zehan Qi, et al. 2023. Survey on factuality in large language models: Knowledge, retrieval and domain-specificity. arXiv preprint arXiv:2310.07521. *Figure 3: Training loss over t…
  50. Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Anjana Arunkumar, Arjun Ashok, Arut Selvan Dhanasekaran, Atharva Naik, David Stap, et al. 2022. Super-naturalinstructions: Generalization via declarative instructions on 1600+ nlp tasks. arXiv preprint arXiv:2204.… find
  51. Chaoyi Wu, Xiaoman Zhang, Ya Zhang, Yanfeng Wang, and Weidi Xie. 2023a. Pmc-llama: Further fine-tuning llama on medical papers. arXiv preprint arXiv:2304.14454.
  52. Shijie Wu, Ozan Irsoy, Steven Lu, Vadim Dabravolski, Mark Dredze, Sebastian Gehrmann, Prabhanjan Kambadur, David Rosenberg, and Gideon Mann. 2023b. Bloomberggpt: A large language model for finance. arXiv preprint arXiv:2303.17564.
  53. Shitao Xiao, Zheng Liu, Peitian Zhang, and Niklas Muennighoff. 2023. C-pack: Packaged resources to advance general chinese embedding. Preprint, arXiv:2309.07597.
  54. Hongyang Yang, Xiao-Yang Liu, and Christina Dan Wang. 2023. Fingpt: Open-source financial large language models. arXiv preprint arXiv:2306.06031. *Figure 4: Model accuracy on the current events task as a function of the number of paraphrases.*

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ovadia-2024-fine-tuning-or-retrieval-comparing/