The DOI matches its Crossref record; 36 of the 39 references it cites were found in a registry or an index.
Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agents
Open accessAbstract
Large Language Models (LLMs) have demonstrated remarkable zero-shot generalization across various language-related tasks, including search engines.However, existing work utilizes the generative ability of LLMs for Information Retrieval (IR) rather than direct passage ranking.The discrepancy between the pretraining objectives of LLMs and the ranking objective poses another challenge.In this paper, we first investigate generative LLMs such as ChatGPT and GPT-4 for relevance ranking in IR.Surprisingly, our experiments reveal that properly instructed LLMs can deliver competitive, even superior results to state-of-the-art supervised methods on popular IR benchmarks.Furthermore, to address concerns about data contamination of LLMs, we collect a new test set called NovelEval, based on the latest knowledge and aiming to verify the model's ability to rank unknown knowledge.Finally, to improve efficiency in real-world applications, we delve into the potential for distilling the ranking capabilities of ChatGPT into small specialized models using a permutation distillation scheme.Our evaluation results turn out that a distilled 440M model outperforms a 3B supervised model on the BEIR benchmark.The code to reproduce our results is available at www.github.com/sunnweiwei/RankGPT.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Rank-biased overlap (RBO)foot_6 is a widely used metric for evaluating the similarity between two ranked lists and is particularly relevant for tasks in information retrieval (Oh et al., 2022;Sun et al., 2023).”
From Benchmarking Cognitive Biases in Large Language Models as Evaluators (2024) · cites this paper as Sun et al., 2023
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“A representative empirical anchor is Sun et al. (2023), who show that LLMs can match or exceed strong supervised re-rankers on standard IR benchmarks, validating the 'generative engine as LLM re-ranker' abstraction we build on.”
From E-GEO: A Testbed for Generative Engine Optimization in E-Commerce (2025) · cites this paper as Sun et al., 2023
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
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602
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
Dense passage retrieval for open-domain question answering
Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., & Yih, W.-T. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.550 · Cited by 3,575
Learning to rank using gradient descent
Burges, C., Shaked, T., Renshaw, E., Lazier, A., Deeds, M., Hamilton, N., & Hullender, G. (2005) · Proceedings of the 22nd international conference on Machine learning - ICML '05 · doi:10.1145/1102351.1102363 · Cited by 2,836
MS MARCO: A Human Generated MAchine Reading COmprehension Dataset
Bajaj, P., Campos, D., Craswell, N., Deng, L., Gao, J., Liu, X., Majumder, R., McNamara, A., Mitra, B., Nguyen, T., et al. (2016) · arXiv · doi:10.48550/arxiv.1611.09268 · Cited by 2,473
Scaling Instruction-Finetuned Language Models
Chung, H. W., Hou, L., Longpre, S., Zoph, B., Tay, Y., Fedus, W., Li, Y., Wang, X., Dehghani, M., Brahma, S., et al. (2022) · arXiv · doi:10.48550/arxiv.2210.11416 · Cited by 1,441
Document Ranking with a Pretrained Sequence-to-Sequence Model
Nogueira, R. F., Jiang, Z., Pradeep, R., & Lin, J. (2020) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2020.findings-emnlp.63 · Cited by 686
Nogueira, R., & Cho, K. (2019) · arXiv · doi:10.48550/arxiv.1901.04085 · Cited by 615
Holistic Evaluation of Language Models
Bommasani, R., Liang, P., & Lee, T. (2023) · Annals of the New York Academy of Sciences · doi:10.1111/nyas.15007 · Cited by 480
Research works printed but not linked: 5
- TMLR.
- Gautier Izacard, Patrick Lewis, Maria Lomeli, Lucas Hosseini, Fabio Petroni, Timo Schick, Jane A. Yu, Armand Joulin, Sebastian Riedel, and Edouard Grave. 2023. Few-shot learning with retrieval augmented language models. Journal of Machine Learning Re- search, 24(251):1-43.
- Nandan Thakur, Nils Reimers, Andreas Ruckl'e, Ab- hishek Srivastava, and Iryna Gurevych. 2021. Beir: A heterogenous benchmark for zero-shot evaluation of information retrieval models. In NeurIPS 2021.
- Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, and Hannaneh Hajishirzi. 2023b. Self-instruct: Aligning language model with self generated instructions. In ACL 2023.
- Lixin Zou, Shengqiang Zhang, Hengyi Cai, Dehong Ma, Suqi Cheng, Daiting Shi, Shuaiqiang Wang, Zhicong Cheng, and Dawei Yin. 2021. Pre-trained language model based ranking in baidu search. In KDD 2021. Method DL19 DL20 Covid NFCorpus Touche DBPedia SciFact Signal News Robust04 BEIR (Avg) BM25 50.58 47.96 59.47 30.75 44.22 31.80 67.89 33.05 39.52 40.70 43.42 Supervised train on MS MRACO monoBERT (340M) 70.50 67.28 70.01 36.88 31.75 41.87 71.36 31.44 44.62 49.35 47.16 monoT5 (220M) 71.48 66.99 78.34 37.38 30.82 42.42 73.40 31.67 46.83 51.72 49.07 monoT5 (3B) 71.83 68.89 80.71 38.97 32.41 44.45 76.57 32.55 48.49 56.71 51.36 Cohere Rerank-v2 73.22 67.08 81.81 36.36 32.51 42.51 74.44 29.60 47.59 50.78 49.45 Unsupervised instructional permutation generation ChatGPT 65.80 62.91 76.67 35.62 36.18 44.47 70.43 32.12 48.85 50.62 49.37 GPT-4 75.59 70.56 85.51 38.47 38.57 47.12 74.95 34.40 52.89 57.55 53.68 Specialized Models train on MARCO labels or ChatGPT predicted permutations MARCO Pointwise BCE 65.57 56.72 70.82 33.10 17.08 32.28 55.37 19.30 41.52 46.00 39.43 MARCO Listwise CE 65.99 57.97 66.31 32.61 20.15 30.79 37.57 18.09 38.11 39.93 35.45 MARCO RankNet 66.34 58.51 70.29 34.23 20.27 29.62 49.01 23.22 39.82 43.87 38.79 MARCO LambdaLoss 64.82 56.16 72.86 34.20 19.51 32.55 51.88 26.22 42.47 45.28 40.62 ChatGPT Listwise CE 65.39 58.80 76.29 35.73 38.19 40.24 64.49 31.37 47.61 48.00 47.74 ChatGPT RankNet 65.75 59.34 81.26 36.57 39.03 42.10 68.77 31.55 52.54 52.44 50.53 ChatGPT LambdaLoss 67.17 60.56 80.63 36.74 36.73 43.75 68.21 32.58 49.00 50.51 49.77 deberta-v3-xsmall (70M) 64.75 55.07 78.21 35.95 35.42 41.37 67.86 30.04 47.68 49.91 48.31 deberta-v3-small (142M) 67.85 58.84 78.88 36.55 36.16 40.99 66.66 30.29 49.17 49.73 48.55 deberta-v3-base (184M) 70.28 62.52 80.81 36.15 37.25 44.06 71.70 32.45 50.84 51.33 50.57 deberta-v3-large (435M) 70.66 67.15 84.64 38.48 39.27 47.36 74.18 32.53 51.19 56.55 53.03 deberta-v3-large 5K 70.93 64.32 84.43 38.66 40.72 46.28 73.88 31.93 52.24 55.89 53.00 deberta-v3-large 3K 70.79 63.91 84.21 38.73 39.83 45.74 74.41 31.92 52.29 57.42 53.07 deberta-v3-large 1K 69.90 64.81 83.38 38.94 36.65 44.46 71.96 30.19 50.73 53.74 51.26 deberta-v3-large 500 69.71 62.00 83.54 37.23 33.68 44.56 70.48 28.70 45.64 42.67 48.31 deberta-v3-large label 10K 66.61 57.26 74.36 33.94 18.09 34.95 35.35 21.38 39.00 44.94 37.75 deberta-v3-large label 5K 68.98 61.38 80.73 35.68 20.48 37.34 54.63 24.25 36.94 51.13 42.64 deberta-v3-large label 3K 67.41 60.42 79.82 35.49 24.54 37.39 47.31 23.29 39.87 50.65 42.29 deberta-v3-large label 1K 65.55 60.93 77.70 33.29 23.36 36.38 31.10 21.71 34.28 38.31 37.01 deberta-v3-large label 500 60.59 54.45 76.20 32.93 19.66 31.54 45.66 13.99 33.48 44.49 37.24 deberta-v3-large monoT5-3B 73.05 68.82 84.78 38.55 34.43 43.61 75.45 30.75 49.85 56.80 51.78 deberta-v3-large chatgpt+label 72.42 67.30 85.96 38.75 35.06 45.43 71.81 28.52 45.91 55.57 50.88 deberta-v3-base label 10k 65.66 59.84 71.63 34.65 16.53 32.59 34.65 22.64 37.60 44.02 36.79 deberta-v3-small label 10k 63.63 52.83 68.17 30.48 18.12 31.72 33.62 18.02 34.57 36.09 33.85 deberta-v3-xsmall label 10k 60.89 51.15 63.58 28.67 14.87 27.12 20.60 18.97 32.61 32.67 29.89 llama-7b 71.33 66.06 78.23 37.60 34.87 45.46 76.13 34.17 51.79 55.22 51.68 vicuna-7b 71.80 66.89 78.32 36.87 31.81 45.40 74.23 34.28 51.13 52.91 50.62 llama-7b 10k label 65.22 56.85 75.36 36.24 20.88 37.34 69.04 25.22 41.21 49.21 44.31 llama-7b 5k label 69.24 58.97 80.49 37.55 28.23 39.66 71.79 26.04 44.09 53.83 47.71
Cited by
Sorted by citations
A Survey on Evaluation of Large Language Models
Chang, Y., Wang, X., Wang, J., Wu, Y., Yang, L., Zhu, K., Chen, H., Yi, X., Wang, C., Wang, Y., et al. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3641289 · Cited by 4,254
A survey on large language models for recommendation
Wu, L., Zheng, Z., Qiu, Z., Wang, H., Gu, H., Shen, T., Qin, C., Zhu, C., Zhu, H., Liu, Q., et al. (2024) · World Wide Web · doi:10.1007/s11280-024-01291-2 · Cited by 515
How Can Recommender Systems Benefit from Large Language Models: A Survey
Lin, J., Dai, X., Xi, Y., Liu, W., Chen, B., Zhang, H., Liu, Y., Wu, C., Li, X., Zhu, C., et al. (2025) · ACM Transactions on Information Systems · doi:10.1145/3678004 · Cited by 242
Large Language Models for Information Retrieval: A Survey
Zhu, Y., Yuan, H., Wang, S., Liu, J., Liu, W., Deng, C., Chen, H., Liu, Z., Dou, Z., & Wen, J.-R. (2026) · ACM Transactions on Information Systems · doi:10.1145/3748304 · Cited by 225
Self-citationDense Text Retrieval Based on Pretrained Language Models: A Survey
Zhao, W. X., Liu, J., Ren, R., & Wen, J.-R. (2024) · ACM Transactions on Information Systems · doi:10.1145/3637870 · Cited by 211
Metadata
- Authors
- Sun, W., Yan, L., Ma, X., Wang, S., Ren, P., Chen, Z., Yin, D., & Ren, Z.
- Affiliations
- Shandong University (CN)
- Baidu (China) (CN)
- Leiden University (NL)
- Venue
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Published
- 2023
- Pages
- 14918-14937
- OpenAlex
- W4389523765
- Cited by
- 270
- References
- 39
- Citations by year
- 2026: 94 · 2025: 104 · 2024: 56 · 2023: 16
- FWCI
- 69.39
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- Leiden Repository (Leiden University) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- Leiden Repository (Leiden University) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- Shandong University
- National Natural Science Foundation of China (61972234, 62102234, 62272274, 62202271, 62072279)
- China Scholarship Council
- Natural Science Foundation of Shandong Province (ZR2021QF129)
- Topics
- Physical Sciences › Computer Science › Information Systems › Information Retrieval and Search Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
- Keywords
- large language models, ChatGPT, GPT-4, re-ranking, information retrieval, passage ranking, relevance ranking, contaminated data, knowledge distillation, BEIR benchmark
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
