Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Cited by 41

The DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.

A fine-tuning enhanced RAG system with quantized influence measure as AI judge

Rangan, Keshav, Yin, Yiqiao · 2024 · Scientific Reports

Abstract

This study presents an innovative enhancement to retrieval-augmented generation (RAG) systems by seamlessly integrating fine-tuned large language models (LLMs) with vector databases. This integration capitalizes on the combined strengths of structured data retrieval and the nuanced comprehension provided by advanced LLMs. Central to our approach are the LoRA and QLoRA methodologies, which stand at the forefront of model refinement through parameter-efficient fine-tuning and memory optimization. A novel feature of our research is the incorporation of user feedback directly into the training process, ensuring the model's continuous adaptation to user expectations and thus, improving its performance and applicability. Additionally, we introduce a Quantized Influence Measure (QIM) as an innovative "AI Judge" mechanism to enhance the precision of result selection, further refining the system's accuracy. Accompanied by an executive diagram and a detailed algorithm for fine-tuning QLoRA, our work provides a comprehensive framework for implementing these advancements within chatbot technologies. This research contributes significant insights into LLM optimization for specific uses and heralds new directions for further development in retrieval-augmented models. Through extensive experimentation and analysis, our findings lay a robust foundation for future advancements in chatbot technology and retrieval systems, marking a significant step forward in the creation of more sophisticated, precise, and user-centric conversational AI systems. We make the dataset , the data processing package huggify-data , the model , and the app publicly available for the community.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-024-79110-x

Cites

12 of 12 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 57

  1. Thompson, S. J., Pollio, D. E., Constantine, J., Reid, D. & Nebbitt, V. Short-term outcomes for youth receiving runaway and homeless shelter services. Res. Soc. Work. Pract. 12(5), 589–603 (2002). · DOI
  2. Spiegler, J., Güereca, C., McQuerry, D., & Troedson, E. From crisis to housing: a comparison of select homeless shelters from across the United States. J. Poverty 28(2), 73–90 (2024). · doi:10.1080/10875549.2022.2113593
  3. Barber, C. C., Fonagy, P., Fultz, J., Simulinas, M. A. & Yates, M. Homeless near a thousand homes: Outcomes of homeless youth in a crisis shelter. Am. J. Orthopsychiatry 75(3), 347–355 (2005). · DOI
  4. Dalton, M. M. & Pakenham, K. I. Adjustment of homeless adolescents to a crisis shelter: Application of a stress and coping model. J. Youth Adolesc. 31, 79–89 (2002). · DOI
  5. Dreyer, B. P. A shelter is not a home: The crisis of family homelessness in the United States. Pediatrics 142(5), e20182695 (2018). · doi:10.1542/peds.2018-2695
  6. Wallace, B., Barber, K. & Pauly, B. B. Sheltering risks: Implementation of harm reduction in homeless shelters during an overdose emergency. Int. J. Drug Policy 53, 83–89 (2018). · DOI
  7. Hurtubise, R., Babin, P.-O. & Grimard, C. Shelters for the homeless: Learning from research. In Finding Home: Policy Options for Addressing Homelessness in Canada (eds Hulchanski, J. D. et al.) 1–24 (Cities Centre, University of Toronto, Toronto, 2009).
  8. Wusinich, C., Bond, L., Nathanson, A. & Padgett, D. K. “if you’re gonna help me, help me’’: Barriers to housing among unsheltered homeless adults. Eval. Program Plan. 76, 101673 (2019). · DOI
  9. Hocking, J. E. & Lawrence, S. G. Changing attitudes toward the homeless: The effects of prosocial communication with the homeless. J. Soc. Distress Homeless 9, 91–110 (2000). · DOI
  10. Brown, M. et al. Waiting for shelter: Perspectives on a homeless shelter’s procedures. J. Commun. Psychol. 45(7), 846–858 (2017). · DOI
  11. Ryan Greysen, S., Allen, R., Lucas, G. I., Wang, E. A. & Rosenthal, M. S. Understanding transitions in care from hospital to homeless shelter: A mixed-methods, community-based participatory approach. J. Gen. Intern. Med. 27, 1484–1491 (2012). · DOI
  12. Vellozzi-Averhoff, C. et al. Disparities in communication among the inpatient homeless population at a safety-net hospital. J. Natl. Med. Assoc. 113(4), 440–448 (2021).
  13. Barker, R. L. At home with the homeless: An experience in transcultural communication. J. Indep. Soc. Work 4(4), 61–73 (1990). · doi:10.1300/j283v04n04_06
  14. Haag, M., Wood, T. & Holloway, L. Impacting quality of life at a homeless shelter: Measuring the effectiveness of say it straight. Int. J. Interdiscip. Soc. Sci. 5(12), 195–204 (2011).
  15. Olufemi, O. Barriers that disconnect homeless people and make homelessness difficult to interpret. Dev. S. Afr. 19(4), 455–466 (2002). · doi:10.1080/0376835022000019455
  16. Haupt, B. B. & Sweeting, K. D. Examining communication for homeless populations in times of crises. Nat. Hazards Rev. 24(3), 05023003 (2023). · doi:10.1061/nhrefo.nheng-1651
  17. He, Z., Xie, Z., Jha, R., Steck, H., Liang, D., Feng, Y., Majumder, B. P., Kallus, N. & McAuley, J. Large language models as zero-shot conversational recommenders. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management 720–730 (2023). · DOI
  18. Babaei Giglou, H., D’Souza, J. & Auer, S. LLMs4OL: Large language models for ontology learning. In The Semantic Web—ISWC 2023: 22nd International Semantic Web Conference, Athens, Greece, November 6–10, 2023, Proceedings, Part I (eds Payne, Terry R. et al.) 408–427 (Springer, Cham, 2023). https://doi.org/10.1007/978-3-031-47240-4_22. · DOI
  19. Ferber, D. & Kather, J. N. Large language models in uro-oncology. Eur. Urol. Oncol. 7(1), 157–159 (2024). · DOI
  20. Ozdemir, S. Quick Start Guide to Large Language Models: Strategies and Best Practices for Using ChatGPT and Other LLMs (Addison-Wesley Professional, Boston, 2023).
  21. Pan, S., Zheng, Y. & Liu, Y. Integrating graphs with large language models: Methods and prospects. arXiv preprint[SPACE]arXiv:2310.05499 (2023). · arXiv
  22. Kumar, V., Srivastava, P., Dwivedi, A., Budhiraja, I., Ghosh, D., Goyal, V. & Arora, R. Large-language-models (llm)-based ai chatbots: Architecture, in-depth analysis and their performance evaluation. In International Conference on Recent Trends in Image Processing and Pattern Recognition 237–249. (Springer 2023). · DOI
  23. Rasnayaka, S., Wang, G., Shariffdeen, R. & Iyer, G. N. An empirical study on usage and perceptions of llms in a software engineering project. arXiv preprint[SPACE]arXiv:2401.16186 (2024). · DOI
  24. Levy, M., Ravfogel, S. & Goldberg, Y. Guiding llm to fool itself: Automatically manipulating machine reading comprehension shortcut triggers. arXiv preprint[SPACE]arXiv:2310.18360 (2023). · DOI
  25. Deng, Z., Gao, H., Miao, Y. & Zhang, H.. Efficient detection of llm-generated texts with a Bayesian surrogate model. arXiv preprint[SPACE]arXiv:2305.16617 (2023). · arXiv
  26. Ge, Y., Hua, W., Ji, J., Tan, J., Xu, S. & Zhang, Y. Openagi: When llm meets domain experts. arXiv preprint[SPACE]arXiv:2304.04370 (2023). · arXiv
  27. Bekbayev, A., Chun, S., Dulat, Y. & Yamazaki, J. The poison of alignment. arXiv preprint[SPACE]arXiv:2308.13449 (2023). · arXiv
  28. Dettmers, T., Pagnoni, A., Holtzman, A. & Zettlemoyer, L. Qlora: Efficient finetuning of quantized llms. arXiv preprint[SPACE]arXiv:2305.14314 (2023). · arXiv
  29. Li, Y., Yu, Y., Liang, C., He, P., Karampatziakis, N., Chen, W. & Zhao, T. Loftq: Lora-fine-tuning-aware quantization for large language models. arXiv preprint[SPACE]arXiv:2310.08659 (2023). · arXiv
  30. Zhang, Z., Zhao, D., Miao, X., Oliaro, G., Li, Q., Jiang, Y. & Jia, Z. Quantized side tuning: Fast and memory-efficient tuning of quantized large language models. arXiv preprint[SPACE]arXiv:2401.07159 (2024). · DOI
  31. Jeon, H., Kim, Y., & Kim, J.-j. L4q: Parameter efficient quantization-aware training on large language models via lora-wise lsq. arXiv preprint[SPACE]arXiv:2402.04902 (2024). · arXiv
  32. Yin, J., Dong, J., Wang, Y., De Sa, C. & Kuleshov, V. Modulora: Finetuning 3-bit llms on consumer gpus by integrating with modular quantizers. arXiv preprint[SPACE]arXiv:2309.16119 (2023). · arXiv
  33. Zhang, X., Rajabi, N., Duh, K. & Koehn, P. Machine translation with large language models: Prompting, few-shot learning, and fine-tuning with qlora. In Proceedings of the Eighth Conference on Machine Translation 468–481 (2023). · DOI
  34. Xu, Y., Xie, L., Gu, X., Chen, X., Chang, H., Zhang, H., Chen, Z., Zhang, X. & Tian, Q. Qa-lora: Quantization-aware low-rank adaptation of large language models. arXiv preprint[SPACE]arXiv:2309.14717 (2023). · arXiv
  35. Guo, H., Greengard, P., Xing, E. P & Kim, Y. Lq-lora: Low-rank plus quantized matrix decomposition for efficient language model finetuning. arXiv preprint[SPACE]arXiv:2311.12023 (2023). · arXiv
  36. Weng, Y., Wang, Z., Liao, H., He, S., Liu, S., Liu, K. & Zhao, J. Lmtuner: An user-friendly and highly-integrable training framework for fine-tuning large language models. arXiv preprint[SPACE]arXiv:2308.10252 (2023). · arXiv
  37. Cai, D., Wang, Y., Liu, L., & Shi, S. Recent advances in retrieval-augmented text generation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval 3417–3419 (2022). · DOI
  38. Liu, S., Chen, Y., Xie, X., Siow, J. & Liu, Y. Retrieval-augmented generation for code summarization via hybrid gnn. arXiv preprint[SPACE]arXiv:2006.05405 (2020). · arXiv
  39. Kim, J., Choi, S., Amplayo, R. K. & Hwang, S.-w. Retrieval-augmented controllable review generation. In Proceedings of the 28th International Conference on Computational Linguistics 2284–2295 (2020). · DOI
  40. Li, H., Su, Y., Cai, D., Wang, Y. & Liu, L. A survey on retrieval-augmented text generation. arXiv preprint[SPACE]arXiv:2202.01110 (2022). · arXiv
  41. Goyal, A. et al. Retrieval-augmented reinforcement learning. In International Conference on Machine Learning 7740–7765 (PMLR, 2022).
  42. Blattmann, A., Rombach, R., Oktay, K., Müller, J. & Ommer, B. Retrieval-augmented diffusion models. Adv. Neural. Inf. Process. Syst. 35, 15309–15324 (2022).
  43. Gao, Y. et al. Retrieval-augmented multilingual keyphrase generation with retriever-generator iterative training. arXiv preprint[SPACE]arXiv:2205.10471 (2022). · DOI · Paper page
  44. Guo, Y., Qiu, W., Leroy, G., Wang, S. & Cohen, T. Retrieval augmentation of large language models for lay language generation. J. Biomed. Inform. 149, 104580 (2024). · DOI
  45. Chernoff, H., Lo, S.-H. & Zheng, T. Discovering influential variables: A method of partitions. Ann. Appl. Stat. 3(4), 1335–1369 (2009). · DOI
  46. Lo, S. H. & Zheng, T. Backward haplotype transmission association algorithm—a fast multiple-marker screening method. Hum. Hered. 53(4), 197–215 (2002). · DOI
  47. Lo, S.-H. & Yin, Y. An interaction-based convolutional neural network (icnn) toward a better understanding of covid-19 x-ray images. Algorithms 14(11), 337 (2021). · DOI
  48. Lo, S.-H. & Yin, Y. A novel interaction-based methodology towards explainable Ai with better understanding of pneumonia chest x-ray images. Discov. Artif. Intell. 1(1), 16 (2021). · DOI
  49. Lo, S.-H. & Yin, Y. Language semantics interpretation with an interaction-based recurrent neural network. Mach. Learn. Knowl. Extr. 3(4), 922–945 (2021). · DOI
  50. Di, X. et al. Detecting mild cognitive impairment and dementia in older adults using naturalistic driving data and interaction-based classification from influence score. Artif. Intell. Med. 138, 102510 (2023). · DOI
  51. Lo, A., Chernoff, H., Zheng, T. & Lo, S.-H. Why significant variables aren’t automatically good predictors. Proc. Natl. Acad. Sci. 112(45), 13892–13897 (2015). · DOI · Paper page
  52. Lo, A., Chernoff, H., Zheng, T. & Lo, S.-H. Framework for making better predictions by directly estimating variables’ predictivity. Proc. Natl. Acad. Sci. 113(50), 14277–14282 (2016). · DOI
  53. Aghajanyan, A., Zettlemoyer, L. & Gupta, S. Intrinsic dimensionality explains the effectiveness of language model fine-tuning. arXiv preprint[SPACE]arXiv:2012.13255 (2020). · DOI · Paper page
  54. He, Y., Liu, J., Wu, W., Zhou, H. & Zhuang, B. Efficientdm: Efficient quantization-aware fine-tuning of low-bit diffusion models. arXiv preprint[SPACE]arXiv:2310.03270 (2023). · arXiv
  55. Schreiber, A. Esmbind and qbind: Lora, qlora, and esm-2 for predicting binding sites and post translational modification. bioRxiv, 2023–11 (2023). · DOI
  56. Zi, B., Qi, X., Wang, L., Wang, J., Wong, K.-F. & Zhang, L. Delta-lora: Fine-tuning high-rank parameters with the delta of low-rank matrices. arXiv preprint[SPACE]arXiv:2309.02411 (2023). · arXiv
  57. Xia, W., Qin, C. & Hazan, E. Chain of lora: Efficient fine-tuning of language models via residual learning. arXiv preprint[SPACE]arXiv:2401.04151 (2024). · arXiv

Cited by

See all 41

Metadata

Authors
Rangan, K., & Yin, Y.
Venue
Scientific Reports
Published
2024-11-10
Volume
14
Issue
1
ISSN
2045-2322
Cited by
41
References
12
Language
en
Licence

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.