Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generation
Open accessThe DOI matches its Crossref record; 76 of the 76 references it cites were found in a registry or an index.
Abstract
Natural language generation has witnessed significant advancements due to the training of large language models on vast internet-scale datasets. Despite these advancements, there exists a critical challenge: These models can inadvertently generate content that is toxic, inaccurate, and unhelpful, and existing automatic evaluation metrics often fall short of identifying these shortcomings. As models become more capable, human feedback is an invaluable signal for evaluating and improving models. This survey aims to provide an overview of recent research that has leveraged human feedback to improve natural language generation. First, we introduce a taxonomy distilled from existing research to categorize and organize the varied forms of feedback. Next, we discuss how feedback can be described by its format and objective, and cover the two approaches proposed to use feedback (either for training or decoding): directly using feedback or training feedback models. We also discuss existing datasets for human-feedback data collection, and concerns surrounding feedback collection. Finally, we provide an overview of the nascent field of AI feedback, which uses large language models to make judgments based on a set of principles and minimize the need for human intervention. We also release a website of this survey at feedback-gap-survey.info.
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.
Integrity signals
Registry
Cites
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.
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
Proximal Policy Optimization Algorithms
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017) · arXiv · doi:10.48550/arxiv.1707.06347 · Cited by 10,625
Watkins, C. J. C. H., & Dayan, P. (1992) · Machine Learning · doi:10.1007/bf00992698 · Cited by 9,083
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988
Answering the Call for a Standard Reliability Measure for Coding Data
Hayes, A. F., & Krippendorff, K. (2007) · Communication Methods and Measures · doi:10.1080/19312450709336664 · Cited by 4,225 · cited 1×
“Inter-rater reliability metrics, such as Cohen’s Kappa, Fleiss’ Kappa, or Krippendorff’s alpha, can assess anno-tator agreement (Hayes and Krippendorff, 2007; Fleiss, 1971; Cohen, 1960).”
Training language models to follow instructions with human feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.02155 · Cited by 4,208
On the Opportunities and Risks of Foundation Models
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., et al. (2021) · arXiv · doi:10.48550/arxiv.2108.07258 · Cited by 2,676
Self-Refine: Iterative Refinement with Self-Feedback
Madaan, A., Tandon, N., Gupta, P., Hallinan, S., Gao, L., Wiegreffe, S., Alon, U., Dziri, N., Prabhumoye, S., Yang, Y., et al. (2023) · arXiv · doi:10.48550/arxiv.2303.17651 · Cited by 2,314
Self-citationCheap and fast---but is it good?: evaluating non-expert annotations for natural language tasks
Snow, R., O'Connor, B., Jurafsky, D., & Ng, A. Y. (2008) · Proceedings of the Conference on Empirical Methods in Natural Language Processing - EMNLP '08 · doi:10.3115/1613715.1613751 · Cited by 1,887
Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V., & Kalai, A. (2016) · arXiv · doi:10.48550/arxiv.1607.06520 · Cited by 1,615 · cited 1×
“However, the data used to train these models is generally scraped from the Internet, often containing noise, social biases, and errors (Bolukbasi et al., 2016; Dodge et al., 2021).”
Concrete Problems in AI Safety
Amodei, D., Olah, C., Steinhardt, J., Christiano, P., Schulman, J., & Mané, D. (2016) · arXiv · doi:10.48550/arxiv.1606.06565 · Cited by 1,314 · cited 1×
“This alignment objective has been studied extensively in the AI safety and alignment literature (Bostrom, 2014; Amodei et al., 2016; Bommasani et al., 2021; Kenton et al., 2021), with Leike et al. (2018) proposing the use of feed-back models to tackle the difficulty in specifying objectives.”
Cited by
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
Zhang, Y., Li, Y., Cui, L., Cai, D., Liu, L., Fu, T., Huang, X., Zhao, E., Zhang, Y., Chen, Y., et al. (2025) · Computational Linguistics · doi:10.1162/coli.a.16 · Cited by 534
Pan, L., Saxon, M., Xu, W., Nathani, D., Wang, X., & Wang, W. Y. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00660 · Cited by 86
A Survey of Reinforcement Learning from Human Feedback
Kaufmann, T., Weng, P., Bengs, V., & Hüllermeier, E. (2023) · arXiv · doi:10.48550/arxiv.2312.14925 · Cited by 43
RL4F: Generating natural language feedback with reinforcement learning for repairing model outputs
Akyurek, A. F., Akyurek, E., Kalyan, A., Clark, P., Wijaya, D. T., & Tandon, N. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.427 · Cited by 29
Feedback by Design: Understanding and Overcoming User Feedback Barriers in Conversational Agents
Sharma, N., Zhang, Z., Lee, D., Krishnan, N., Ren, G.-J., Xiao, Z., & Li, Y. (2026) · Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3772318.3791600 · Cited by 21
Human-inspired Perspectives: A Survey on AI Long-term Memory
He, Z., Lin, W., Zheng, H., Zhang, F., Jones, M. W., Aitchison, L., Xu, X., Liu, M., Kristensson, P. O., & Shen, J. (2024) · arXiv · doi:10.48550/arxiv.2411.00489 · Cited by 3
Metadata
- Authors
- Fernandes, P., Madaan, A., Liu, E., Farinhas, A., Martins, P. H., Bertsch, A., de Souza, J. G. C., Zhou, S., Wu, T., Neubig, G., & Martins, A. F. T.
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2023
- Volume
- 11
- Pages
- 1643-1668
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W4390035079
- Cited by
- 44
- References
- 76
- FWCI
- 3.87
- Type
- journal article
- Language
- en
- Open access
- diamond
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.