Digital forgetting in large language models: a survey of unlearning methods
Open accessThe DOI matches its Crossref record; 131 of the 134 references it cites were found in a registry or an index; 62 of the 114 sources its claims cite were checked for support.
Abstract
Abstract Large language models (LLMs) have become the state of the art in natural language processing. The massive adoption of generative LLMs and the capabilities they have shown have prompted public concerns regarding their impact on the labor market, privacy, the use of copyrighted work, and how these models align with human ethics and the rule of law. As a response, new regulations are being pushed, which require developers and service providers to evaluate, monitor, and forestall or at least mitigate the risks posed by their models. One mitigation strategy is digital forgetting: given a model with undesirable knowledge or behavior, the goal is to obtain a new model where the detected issues are no longer present. Digital forgetting is usually enforced via machine unlearning techniques, which modify trained machine learning models for them to behave as models trained on a subset of the original training data. In this work, we describe the motivations and desirable properties of digital forgetting when applied to LLMs, and we survey recent works on machine unlearning. Specifically, we propose a taxonomy of unlearning methods based on the reach and depth of the modifications done on the models, we discuss and compare the effectiveness of machine unlearning methods for LLMs proposed so far, and we survey their evaluation. Finally, we describe open problems of machine unlearning applied to LLMs and we put forward recommendations for developers and practitioners.
Overview
CoChat Summary
## TL;DR
This is a survey of digital forgetting and machine unlearning for large language models (LLMs), with a taxonomy organized by how deeply methods modify the model and a discussion of evaluation and open challenges. It compares 22 unlearning methods, 28 forgetting datasets, 31 retaining datasets, and 15 LLMs (plus size variants), concluding that method choice depends on the forgetting target, guarantees, utility, and runtime.
## Key findings
- The survey analyzes and compares 22 unlearning methods, 28 datasets used for unlearning and evaluation of forgetting, 31 retaining datasets, and 15 models (plus size variations) used in the literature.
- The paper proposes a taxonomy with four main categories of unlearning methods: global weight modification, local weight modification, architecture modification, and input/output modification.
- It reports that runtime for surveyed methods ranges from minutes to hours, and in one case longer than a day; examples given include minutes for Wang et al. (2023), Yao et al. (2023), Limisiewicz and Mareček (2022), Jang et al. (2022), Yu et al. (2023), hours for Chen and Yang (2023), and longer than a day for Lu et al. (2022).
- The paper notes that utility degradation in retained performance usually falls within 1–5%, though it can reach 20% in some studies; it also states that larger models tend to exhibit better utility retention.
## Methods
- Literature survey and synthesis of recent work on digital forgetting and machine unlearning in LLMs, focusing on papers mostly published in 2023 and early 2024.
- The review organizes methods by the depth and reach of the model changes and by their working principles, then compares their forgetting, retaining, and runtime evaluation practices.
- The paper’s evidence base is the set of surveyed works and their reported datasets, models, and metrics; it does not present a new experimental dataset or original model training.
## Limitations
- The paper states that no de facto standard datasets exist to explicitly evaluate unlearning methods, which makes comparisons difficult.
- It notes that most surveyed works provide no provable guarantees, relying instead on empirical evaluation or auditing; this limits ex ante assurance of forgetting.
- It says that many evaluations are dataset-specific, all surveyed methods are evaluated on English-language datasets, and there is no evidence about performance in other languages.
- Evident from the text: this is a survey, so its conclusions depend on the scope and quality of the existing literature rather than new experimental results.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
42 Valid · 5 Not valid · 28 Not checked · 127 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“Most papers (Eldan and Russinovich 2023; Jang et al 2022; Yao et al 2023; Pawelczyk et al 2023; Chen and Yang 2023; Ni et al 2023; Pochinkov and Schoots 2023 ) leverage the predefined performance metric of the forgetting dataset, that is, the accuracy in a classification or completion dataset.”
What the sources say
- Partly supportedcites Eldan and Russinovich 2023 matched to “Who's Harry Potter? Approximate Unlearning in LLMs” (2023) sourceExperimental — accuracy not yet measured
Section 3.2.1 and Additional information (pages not provided) describe evaluating forgetting through completions of 300 prompts. But the paper uses a custom, weighted familiarity score rather than a predefined accuracy metric for a forgetting dataset. Section 3.1 uses established benchmarks to assess preservation of general capabilities, not forgetting. (LLM, conf=0.86)
- Disputedcites Jang et al 2022 matched to “Knowledge Unlearning for Mitigating Privacy Risks in Language Models” (2022) sourceExperimental — accuracy not yet measured
Section 3.2 (p. 5) says the paper introduces Extraction Likelihood and uses it alongside Memorization Accuracy to determine whether target token sequences have been forgotten. The classification-dataset accuracy reported in Table 2 (Section 4.1, p. 7) measures retained general language-model performance, not forgetting. Thus this paper does not use a predefined classification- or completion-dataset accuracy metric as its forgetting measure. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
- Disputedcites Yao et al 2023 matched to “Large Language Model Unlearning” (2023) sourceExperimental — accuracy not yet measured
Section 5 (Evaluation Design) defines context-dependent unlearning efficacy rather than dataset accuracy. In the harmfulness application (Dataset and Model), the paper measures the rate of outputs flagged as harmful; in the copyrighted-content application (Setting), it measures the rate of completions exceeding a BLEU-based leakage threshold. Neither is the predefined accuracy metric attributed to this paper. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Related surveys on digital forgetting and machine unlearning, such as Qu et al (2023); Xu et al (2023); Nguyen et al (2022); Shaik et al (2023), have focused on the general ML case by considering models used to classify tabular or image data.”
What the sources say
- Not matchedcites Qu et al 2023
This doesn’t read as a citation we can look up.
- Not matchedcites Xu et al 2023
This doesn’t read as a citation we can look up.
- Partly supportedcites Nguyen et al 2022 matched to “A Survey of Machine Unlearning” (2022) sourceExperimental — accuracy not yet measured
The Introduction (p. 2) describes this as a comprehensive survey of machine unlearning, and the published-datasets table (pp. 14–16) includes tabular and image classification. But the same table also covers text, sequence, and graph data, so the source does not support characterizing its focus solely as classification of tabular or image data. (LLM, conf=0.84)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, GA can cause the model to lose its understanding of the language (Eldan and Russinovich 2023 ).”
What the sources say
- Supportedcites Eldan and Russinovich 2023 matched to “Who's Harry Potter? Approximate Unlearning in LLMs” (2023) source
In Section 2, the paper describes training on text while negating the loss function and explains that this could cause the model to unlearn the meaning of ordinary words rather than just the target material. It identifies the underlying risk as impairing understanding of language in general, which supports the citing sentence's qualified 'can cause' claim. (LLM, conf=0.90)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“The authors demonstrate the ability of their method to decrease stereotyping in questionanswering over nine different social groups with a single prompt.”
What the sources say
- Supportedcites Gallegos et al (2024) matched to “Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes” (2024) source
The Introduction explicitly says the authors demonstrate reduced stereotyping in question-answering across nine social groups with a single prompt. The Conclusion also reports bias reductions across nine groups. The evaluation was limited to multiple-choice questions, but the citing sentence does not claim otherwise. (LLM, conf=0.95)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Furthermore, the success of unlearning depends on the specific target data and the domain of the to-be-forgotten data (Smith et al 2023).”
What the sources say
- Partly supportedcites Smith et al 2023 matched to “Identifying and Mitigating Privacy Risks Stemming from Language Models: A Survey” (2023) sourceExperimental — accuracy not yet measured
Section 12.2 (Machine Unlearning; page unavailable) says the forgetting threshold depends on which data samples are chosen and reports that larger forget sets degraded model performance. This lends limited support to the importance of target data, but the passage concerns the measurement of forgetting rather than directly establishing that unlearning success depends on the target samples. It does not address whether success depends on the domain of the data to be forgotten. (LLM, conf=0.77)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
47 more claims · 11 with their cited sources checked
Create a free account to see all 202 claims and to verify the 15 references still open.
- Probably not a substantial claim. 96
- It may state two separate claims as one. 86
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 4×
“Namely, encoder-only architectures such as BERT (Devlin et al 2018 ) –used mostly for text classification–, decoder-only archi - tectures such as GPT (Radford et al 2018) – for text generation–, and encoder-decoder architectures, such as T5 (Raffel et al 2020) –for text-to-text processing, such as…”
InfluentialAdvances in neural information processing systems 7
Cios, K., & Shields, M. (1997) · Neurocomputing · doi:10.1016/s0925-2312(97)90018-7 · Cited by 22,059
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019) · arXiv · doi:10.48550/arxiv.1907.11692 · Cited by 16,515
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
Socher, R., Perelygin, A., Wu, J., Chuang, J., Manning, C. D., Ng, A., & Potts, C. (2013) · Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d13-1170 · Cited by 6,878
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578 · cited 1×
“Mechanisms such as differentially private stochastic gradient descent (DP-SGD) (Abadi et al 2016) and private aggregation of teacher ensembles (PATE) (Papernot et al 2018) have been used in the last years to train privacy-preserving machine learning models.”
Membership Inference Attacks Against Machine Learning Models
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017) · 2017 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2017.41 · Cited by 4,754 · cited 4×
“Even when no ex ante guarantees exist, evaluation of unlearning for privacy (which amounts to evaluating whether the model knows a particular information) has been studied for years, and techniques such as MIAs can be used for empirical evaluation (Carlini et al 2021; Nasr et al 2023).”
InfluentialDistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019) · arXiv · doi:10.48550/arxiv.1910.01108 · Cited by 4,653
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
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., & Neubig, G. (2023) · ACM Computing Surveys · doi:10.1145/3560815 · Cited by 4,361 · cited 1×
“In this sense, this approach can be seen as an instance of prompt engineering (Liu et al 2023) which involves editing the prompts.”
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., & Neubig, G. (2021) · arXiv · doi:10.48550/arxiv.2107.13586 · Cited by 4,352 · cited 1×
“In this sense, this approach can be seen as an instance of prompt engineering (Liu et al 2023) which involves editing the prompts.”
ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., & Soricut, R. (2019) · arXiv · doi:10.48550/arxiv.1909.11942 · Cited by 4,308
Axiomatic attribution for deep networks
Sundararajan, M., Taly, A., & Yan, Q. (2017) · arXiv · doi:10.48550/arxiv.1703.01365 · Cited by 4,087
Llama 2: Open Foundation and Fine-Tuned Chat Models
Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2307.09288 · Cited by 2,604
Parameter-efficient transfer learning for NLP
Houlsby, N., Giurgiu, A., Jastrzebski, S., Morrone, B., de Laroussilhe, Q., Gesmundo, A., Attariyan, M., & Gelly, S. (2019) · arXiv · doi:10.48550/arxiv.1902.00751 · Cited by 2,270
ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
Clark, K., Luong, M.-T., Le, Q. V., & Manning, C. D. (2020) · arXiv · doi:10.48550/arxiv.2003.10555 · Cited by 1,844
Hierarchical Neural Story Generation
Fan, A., Lewis, M., & Dauphin, Y. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1082 · Cited by 1,266
Hierarchical Neural Story Generation
Fan, A., Lewis, M., & Dauphin, Y. (2018) · arXiv · doi:10.48550/arxiv.1805.04833 · Cited by 1,242
Personalizing Dialogue Agents: I have a dog, do you have pets too?
Zhang, S., Dinan, E., Urbanek, J., Szlam, A., Kiela, D., & Weston, J. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1205 · Cited by 1,177
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
Amini, A., Gabriel, S., Lin, S., Koncel-Kedziorski, R., Choi, Y., & Hajishirzi, H. (2019) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/n19-1245 · Cited by 1,087
BLEURT: Learning Robust Metrics for Text Generation
Sellam, T., Das, D., & Parikh, A. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.704 · Cited by 1,005
BLEURT: Learning Robust Metrics for Text Generation
Sellam, T., Das, D., & Parikh, A. P. (2020) · arXiv · doi:10.48550/arxiv.2004.04696 · Cited by 1,005
Privacy Risk in Machine Learning: Analyzing the Connection to Overfitting
Yeom, S., Giacomelli, I., Fredrikson, M., & Jha, S. (2018) · 2018 IEEE 31st Computer Security Foundations Symposium (CSF) · doi:10.1109/csf.2018.00027 · Cited by 1,001
TruthfulQA: Measuring How Models Mimic Human Falsehoods
Lin, S., Hilton, J., & Evans, O. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.229 · Cited by 904
TruthfulQA: Measuring How Models Mimic Human Falsehoods
Lin, S., Hilton, J., & Evans, O. (2021) · arXiv · doi:10.48550/arxiv.2109.07958 · Cited by 899
Salem, A., Zhang, Y., Humbert, M., Berrang, P., Fritz, M., & Backes, M. (2018) · arXiv · doi:10.48550/arxiv.1806.01246 · Cited by 873
Salem, A., Zhang, Y., Humbert, M., Berrang, P., Fritz, M., & Backes, M. (2019) · Proceedings 2019 Network and Distributed System Security Symposium · doi:10.14722/ndss.2019.23119 · Cited by 852 · cited 1×
“…ML models have been shown to memorize and leak data from the training data-set, and this also holds for LLMs. Membership inference attacks (MIA) (Shokri et al 2017; Salem et al 2018; Carlini et al 2021) seek to determine whether some information was part of the training set of a trained model.”
HellaSwag: Can a Machine Really Finish Your Sentence?
Zellers, R., Holtzman, A., Bisk, Y., Farhadi, A., & Choi, Y. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1472 · Cited by 830 · cited 1×
“Eldan and Russinovich (2023), Wang et al (2023), and Jang et al (2022) leverage well-established LLM benchmarks, such as those used for comparing LLMs (Sakaguchi et al 2020; Zellers et al 2019; Paperno et al 2016; Bisk et al 2020).”
Bourtoule, L., Chandrasekaran, V., Choquette-Choo, C. A., Jia, H., Travers, A., Zhang, B., Lie, D., & Papernot, N. (2021) · 2021 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp40001.2021.00019 · Cited by 810 · cited 3×
“Bourtoule et al (2021) introduce SISA (Sharded, Isolated, Sliced, and Aggregated training) as a generic exact unlearning framework.”
Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset
Rashkin, H., Smith, E. M., Li, M., & Boureau, Y.-L. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1534 · Cited by 771
Textbooks Are All You Need II: phi-1.5 technical reportPreprint
Li, Y.-F., Bubeck, S., Eldan, R., Giorno, A. D., Gunasekar, S., & Lee, Y. (2023) · arXiv.org · doi:10.48550/arxiv.2309.05463 · Cited by 713
RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Gehman, S., Gururangan, S., Sap, M., Choi, Y., & Smith, N. A. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.301 · Cited by 693
PubMedQA: A Dataset for Biomedical Research Question Answering
Jin, Q., Dhingra, B., Liu, Z., Cohen, W., & Lu, X. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1259 · Cited by 685
Gender bias in coreference resolution: Evaluation and debiasing methods
Zhao, J., Wang, T., Yatskar, M., Ordonez, V., & Chang, K.-W. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2003 · Cited by 684
Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection
Ravfogel, S., Elazar, Y., Gonen, H., Twiton, M., & Goldberg, Y. (2020) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2004.07667 · Cited by 655
StereoSet: Measuring stereotypical bias in pretrained language models
Nadeem, M., Bethke, A., & Reddy, S. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.416 · Cited by 643
StereoSet: Measuring stereotypical bias in pretrained language models
Nadeem, M., Bethke, A., & Reddy, S. (2020) · arXiv · doi:10.48550/arxiv.2004.09456 · Cited by 643
Galactica: A Large Language Model for Science
Taylor, R., Kardas, M., Cucurull, G., Scialom, T., Hartshorn, A., Saravia, E., Poulton, A., Kerkez, V., & Stojnic, R. (2022) · arXiv · doi:10.48550/arxiv.2211.09085 · Cited by 605
WinoGrande: An Adversarial Winograd Schema Challenge at Scale
Sakaguchi, K., Le Bras, R., Bhagavatula, C., & Choi, Y. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6399 · Cited by 579
Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model
Üstün, A., Aryabumi, V., Yong, Z., Ko, W.-Y., D’souza, D., Onilude, G., Bhandari, N., Singh, S., Ooi, H.-L., Kayid, A., et al. (2024) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2024.acl-long.845 · Cited by 576
Wizard of Wikipedia: Knowledge-Powered Conversational agents
Dinan, E., Roller, S., Shuster, K., Fan, A., Auli, M., & Weston, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.01241 · Cited by 572
The Pile: An 800GB Dataset of Diverse Text for Language Modeling
Gao, L., Biderman, S., Black, S., Golding, L., Hoppe, T., Foster, C., Phang, J., He, H., Thite, A., Nabeshima, N., et al. (2021) · arXiv · doi:10.48550/arxiv.2101.00027 · Cited by 492
Zero-Shot Relation Extraction via Reading Comprehension
Levy, O., Seo, M., Choi, E., & Zettlemoyer, L. (2017) · Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017) · doi:10.18653/v1/k17-1034 · Cited by 482 · cited 1×
“Note that the zsRE (Levy et al 2017) dataset and those subsequently listed were also employed for forgetting purposes by the same authors (see Table 2).”
Zero-Shot Relation Extraction via Reading Comprehension
Levy, O., Seo, M., Choi, E., & Zettlemoyer, L. (2017) · arXiv · doi:10.48550/arxiv.1706.04115 · Cited by 482 · cited 1×
“Note that the zsRE (Levy et al 2017) dataset and those subsequently listed were also employed for forgetting purposes by the same authors (see Table 2).”
Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
Clark, P., Cowhey, I., Etzioni, O., Khot, T., Sabharwal, A., Schoenick, C., & Tafjord, O. (2018) · arXiv · doi:10.48550/arxiv.1803.05457 · Cited by 462
Aya Model: An Instruction Finetuned Open-Access Multilingual Language ModelPreprint
Ustun, A., Aryabumi, V., Yong, Z.-X., Ko, W.-Y., D'Souza, D., Onilude, G., Bhandari, N., Singh, S., Ooi, H.-L., Kayid, A., et al. (2024) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2402.07827 · Cited by 447
CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models
Nangia, N., Vania, C., Bhalerao, R., & Bowman, S. R. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.154 · Cited by 446
Gender Bias in Coreference Resolution
Rudinger, R., Naradowsky, J., Leonard, B., & Van Durme, B. (2018) · Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers) · doi:10.18653/v1/n18-2002 · Cited by 443
Pointer Sentinel Mixture Models
Merity, S., Xiong, C., Bradbury, J., & Socher, R. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1609.07843 · Cited by 427
SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization
Gliwa, B., Mochol, I., Biesek, M., & Wawer, A. (2019) · Proceedings of the 2nd Workshop on New Frontiers in Summarization · doi:10.18653/v1/d19-5409 · Cited by 424
A Comprehensive Study of Knowledge Editing for Large Language ModelsPreprint
Zhang, N., Yao, Y., Tian, B., Wang, P., Deng, S., Wang, M., Xi, Z., Mao, S., Zhang, J., Ni, Y., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2401.01286 · Cited by 407
Scalable Private Learning with PATE
Papernot, N., Song, S., Mironov, I., Raghunathan, A., Talwar, K., & Erlingsson, Ú. (2018) · arXiv · doi:10.48550/arxiv.1802.08908 · Cited by 373
HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models
Li, J., Cheng, X., Zhao, X., Nie, J.-Y., & Wen, J.-R. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.397 · Cited by 343
HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models
Li, J., Cheng, X., Zhao, W. X., Nie, J.-Y., & Wen, J.-R. (2023) · arXiv · doi:10.48550/arxiv.2305.11747 · Cited by 340
Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting
De-Arteaga, M., Romanov, A., Wallach, H., Chayes, J., Borgs, C., Chouldechova, A., Geyik, S., Kenthapadi, K., & Kalai, A. T. (2019) · Proceedings of the Conference on Fairness, Accountability, and Transparency · doi:10.1145/3287560.3287572 · Cited by 337 · cited 3×
“This is because the objective was to selectively forget a subset of samples, such as specific generations ( i.e. , private information in Enron emails (Klimt and Yang 2004)), a concept ( i.e. , gender bias in Bias in Bios (De-Arteaga et al 2019)) or a task ( e.g. , Python programming in CodeParrot GitHub Code (Tunstal…”
Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting
De-Arteaga, M., Romanov, A., Wallach, H., Chayes, J., Borgs, C., Chouldechova, A., Geyik, S., Kenthapadi, K., & Kalai, A. T. (2019) · arXiv · doi:10.48550/arxiv.1901.09451 · Cited by 337
Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection
Nivre, J., de Marneffe, M.-C., Ginter, F., Hajič, J., Manning, C. D., Pyysalo, S., Schuster, S., Tyers, F. M., & Zeman, D. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.10643 · Cited by 325
Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classification
Borkan, D., Dixon, L., Sorensen, J., Thain, N., & Vasserman, L. (2019) · Companion Proceedings of The 2019 World Wide Web Conference · doi:10.1145/3308560.3317593 · Cited by 309
Evaluating Gender Bias in Machine Translation
Stanovsky, G., Smith, N. A., & Zettlemoyer, L. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1164 · Cited by 292
Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection
Ravfogel, S., Elazar, Y., Gonen, H., Twiton, M., & Goldberg, Y. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.647 · Cited by 287 · cited 1×
“This is because the bias signal can be encoded non-linearly in LLMs and, even when the whole bias subspace is removed, the information can be recovered in the next layer of the model (Ravfogel et al 2020).”
Unlearn What You Want to Forget: Efficient Unlearning for LLMs
Chen, J., & Yang, D.-Y. (2023) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2310.20150 · Cited by 274
Membership Inference Attacks against Machine Learning ModelsPreprint
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.05820 · Cited by 273
United Nations Universal Declaration of Human Rights
(2003) · Rights and Reason · doi:10.1017/upo9781844653461.016 · Cited by 260
Locating and Editing Factual Associations in GPT
Meng, K., Bau, D., Andonian, A., & Belinkov, Y. (2022) · arXiv · doi:10.48550/arxiv.2202.05262 · Cited by 230
Mind the GAP: A balanced corpus of gendered ambiguous pronouns
Webster, K., Recasens, M., Axelrod, V., & Baldridge, J. (2018) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00240 · Cited by 223
A Gold Standard Dependency Corpus for English
Silveira, N., Dozat, T., de Marneffe, M.-C., Bowman, S. R., Connor, M. K., Bauer, J., & Manning, C. D. (2014) · Language Resources and Evaluation Conference (LREC) · doi:10.63317/4esbhdz7y5u5 · Cited by 217
Memory-assisted prompt editing to improve GPT-3 after deployment
Madaan, A., Tandon, N., Clark, P., & Yang, Y.-M. (2022) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2201.06009 · Cited by 213
A Survey of Machine Unlearning
Nguyen, T. T., Huynh, T. T., Ren, Z., Nguyen, P. L., Liew, A. W.-C., Yin, H., & Nguyen, Q. V. H. (2022) · arXiv · doi:10.48550/arxiv.2209.02299 · Cited by 193
Internet-Augmented Dialogue Generation
Komeili, M., Shuster, K., & Weston, J. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.579 · Cited by 192
Can You Put it All Together: Evaluating Conversational Agents’ Ability to Blend Skills
Smith, E. M., Williamson, M., Shuster, K., Weston, J., & Boureau, Y.-L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.183 · Cited by 170
Personalizing Dialogue Agents: I have a dog, do you have pets too?
Zhang, S., Dinan, E., Urbanek, J., Szlam, A., Kiela, D., & Weston, J. (2018) · arXiv · doi:10.48550/arxiv.1801.07243 · Cited by 165
Xu, H., Zhu, T., Zhang, L., Zhou, W., & Yu, P. S. (2024) · ACM Computing Surveys · doi:10.1145/3603620 · Cited by 164
Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
Biderman, S., Schoelkopf, H., Anthony, Q., Bradley, H., O'Brien, K., Hallahan, E., Khan, M. A., Purohit, S., Prashanth, U. S., Raff, E., et al. (2023) · arXiv · doi:10.48550/arxiv.2304.01373 · Cited by 160
Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomy
Shaik, T., Tao, X., Xie, H., Li, L., Zhu, X., & Li, Q. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3486109 · Cited by 159
Unlearning Bias in Language Models by Partitioning Gradients
Yu, C., Jeoung, S., Kasi, A., Yu, P., & Ji, H. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.375 · Cited by 158
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
Patil, V., Hase, P., & Bansal, M. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2309.17410 · Cited by 127
KGA: A General Machine Unlearning Framework Based on Knowledge Gap AlignmentPreprint
Wang, L., Chen, T., Yuan, W., et al. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2305.06535 · Cited by 121
Scalable Extraction of Training Data from (Production) Language Models
Nasr, M., Carlini, N., Hayase, J., Jagielski, M., Cooper, A. F., Ippolito, D., Choquette-Choo, C. A., Wallace, E., Tramèr, F., & Lee, K. (2023) · arXiv · doi:10.48550/arxiv.2311.17035 · Cited by 96 · cited 2×
“Even when no ex ante guarantees exist, evaluation of unlearning for privacy (which amounts to evaluating whether the model knows a particular information) has been studied for years, and techniques such as MIAs can be used for empirical evaluation (Carlini et al 2021; Nasr et al 2023).”
BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset
Ji, J., Liu, M., Dai, J., Pan, X., Zhang, C., Bian, C., Zhang, C., Sun, R., Wang, Y., & Yang, Y. (2023) · arXiv · doi:10.48550/arxiv.2307.04657 · Cited by 88
The LAMBADA dataset: Word prediction requiring a broad discourse context
Paperno, D., Kruszewski, G., Lazaridou, A., Pham, N. Q., Bernardi, R., Pezzelle, S., Baroni, M., Boleda, G., & Fernandez, R. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1144 · Cited by 87 · cited 1×
“Eldan and Russinovich (2023), Wang et al (2023), and Jang et al (2022) leverage well-established LLM benchmarks, such as those used for comparing LLMs (Sakaguchi et al 2020; Zellers et al 2019; Paperno et al 2016; Bisk et al 2020).”
Self-Detoxifying Language Models via Toxification Reversal
Leong, C. T., Cheng, Y., Wang, J., Wang, J., & Li, W. (2023) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2023.emnlp-main.269 · Cited by 81
C‐sanitized: A privacy model for document redaction and sanitization
Sánchez, D., & Batet, M. (2016) · Journal of the Association for Information Science and Technology · doi:10.1002/asi.23363 · Cited by 80 · cited 1×
“More more sophisticated approaches assess the sensitiveness of terms by leveraging the information theory (Sánchez and Batet 2016, 2017), or the semantics embedded in language models (Hassan et al 2019, 2023).”
Self-citationAnonymisation Models for Text Data: State of the art, Challenges and Future Directions
Lison, P., Pilán, I., Sanchez, D., Batet, M., & Øvrelid, L. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.323 · Cited by 80 · cited 1×
“Most approaches are based on named-entity recognition (either rule-based or ML-based) (Lison et al 2021).”
BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset
Ji, J., Liu, M., Dai, J., Pan, X., Zhang, C., Bian, C., Chen, B., Sun, R., Wang, Y., & Yang, Y. (2023) · neural information processing systems · doi:10.52202/075280-1072 · Cited by 70
Deduplicating Training Data Mitigates Privacy Risks in Language Models
Kandpal, N., Wallace, E., & Raffel, C. (2022) · arXiv · doi:10.48550/arxiv.2202.06539 · Cited by 68
Editing Models with Task Arithmetic
Ilharco, G., Ribeiro, M. T., Wortsman, M., Gururangan, S., Schmidt, L., Hajishirzi, H., & Farhadi, A. (2022) · arXiv · doi:10.48550/arxiv.2212.04089 · Cited by 66
MasakhaNER: Named entity recognition for African languages
(2021) · HAL · doi:10.1162/tacl · Cited by 61
Memory-assisted prompt editing to improve GPT-3 after deployment
Madaan, A., Tandon, N., Clark, P., & Yang, Y. (2022) · Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2022.emnlp-main.183 · Cited by 60 · cited 1×
“To correct model errors via user interactions without retraining, MemPrompt (Madaan et al 2022) pairs the model with a growing memory of recorded cases where the model misunderstood the user intent.”
Large Language Model Unlearning
Yao, Y., Xu, X., & Liu, Y. (2023) · arXiv · doi:10.48550/arxiv.2310.10683 · Cited by 53
Selective Forgetting: Advancing Machine Unlearning Techniques and Evaluation in Language ModelsPreprint
Wang, L.-Z., Zeng, X., Guo, J.-S., Wong, K.-F., & Gottlob, G. (2024) · AAAI Conference on Artificial Intelligence · doi:10.48550/arxiv.2402.05813 · Cited by 53
(2024) · doi:10.52202/079017-3346 · Cited by 53
Unlearn What You Want to Forget: Efficient Unlearning for LLMs
Chen, J., & Yang, D. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.738 · Cited by 52
Memory-Based Model Editing at Scale
Mitchell, E., Lin, C., Bosselut, A., Manning, C. D., & Finn, C. (2022) · arXiv · doi:10.48550/arxiv.2206.06520 · Cited by 51
QUARK: Controllable Text Generation with Reinforced Unlearning
Lu, X., Welleck, S., Hessel, J., Jiang, L., Qin, L., West, P., Ammanabrolu, P., & Choi, Y. (2022) · neural information processing systems · doi:10.52202/068431-2001 · Cited by 46
LEDGAR : a large-scale multi-label corpus for text classification of legal provisions in contracts
Tuggener, D., von Däniken, P., Peetz, T., & Cieliebak, M. (2020) · Zürcher Hochschule für Angewandte Wissenschaften digital collection (Zurich University of Applied Sciences) · doi:10.21256/zhaw-20087 · Cited by 46
LEACE: Perfect linear concept erasure in closed form
Belrose, N., Schneider-Joseph, D., Ravfogel, S., Cotterell, R., Raff, E., & Biderman, S. (2023) · arXiv · doi:10.48550/arxiv.2306.03819 · Cited by 44
Utility-Preserving Privacy Protection of Textual Documents via Word Embeddings
Hassan, F., Sánchez, D., & Domingo‐Ferrer, J. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3076632 · Cited by 42
Self-citationDEPN: Detecting and Editing Privacy Neurons in Pretrained Language Models
Wu, X., Li, J., Xu, M., Dong, W., Wu, S., Bian, C., & Xiong, D. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2023.emnlp-main.174 · Cited by 38
Learn to Unlearn: A Survey on Machine UnlearningPreprint
Qu, Y., Yuan, X., Ding, M., Ni, W., Rakotoarivelo, T., & Smith, D. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.07512 · Cited by 38
PIQA: Reasoning about Physical Commonsense in Natural Language
Bisk, Y., Zellers, R., Le bras, R., Gao, J., & Choi, Y. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6239 · Cited by 37 · cited 1×
“Eldan and Russinovich (2023), Wang et al (2023), and Jang et al (2022) leverage well-established LLM benchmarks, such as those used for comparing LLMs (Sakaguchi et al 2020; Zellers et al 2019; Paperno et al 2016; Bisk et al 2020).”
Detecting Pretraining Data from Large Language Models
Shi, W., Ajith, A., Xia, M., Huang, Y., Liu, D., Blevins, T., Chen, D., & Zettlemoyer, L. (2023) · arXiv · doi:10.48550/arxiv.2310.16789 · Cited by 37 · cited 1×
“However, Shi et al (2023) found that models that underwent unlearning with this approach could still output related copyrighted content.”
HellaSwag: Can a Machine Really Finish Your Sentence?
Zellers, R., Holtzman, A., Bisk, Y., Farhadi, A., & Choi, Y. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.07830 · Cited by 35
Automatic Anonymization of Textual Documents: Detecting Sensitive Information via Word Embeddings
Hassan, F., Sanchez, D., Soria-Comas, J., & Domingo-Ferrer, J. (2019) · 2019 18th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/13th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE) · doi:10.1109/trustcom/bigdatase.2019.00055 · Cited by 34 · cited 1×
“More more sophisticated approaches assess the sensitiveness of terms by leveraging the information theory (Sánchez and Batet 2016, 2017), or the semantics embedded in language models (Hassan et al 2019, 2023).”
Self-citationToward sensitive document release with privacy guarantees
Sánchez, D., & Batet, M. (2017) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2016.12.013 · Cited by 33 · cited 1×
“More more sophisticated approaches assess the sensitiveness of terms by leveraging the information theory (Sánchez and Batet 2016, 2017), or the semantics embedded in language models (Hassan et al 2019, 2023).”
Self-citationKGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignment
Wang, L., Chen, T., Yuan, W., Zeng, X., Wong, K.-F., & Yin, H. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.740 · Cited by 25
In-Context Unlearning: Language Models as Few Shot Unlearners
Pawelczyk, M., Neel, S., & Lakkaraju, H. (2023) · arXiv · doi:10.48550/arxiv.2310.07579 · Cited by 20
Who's Harry Potter? Approximate Unlearning in LLMs
Eldan, R., & Russinovich, M. (2023) · arXiv · doi:10.48550/arxiv.2310.02238 · Cited by 19
Ni, S.-W., Chen, D.-W., Li, C.-M., Hu, X.-P., Xu, R.-F., & Yang, M. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2311.08011 · Cited by 19
Limisiewicz, T., & Mareček, D. (2022) · Proceedings of the 4th Workshop on Gender Bias in Natural Language Processing (GeBNLP) · doi:10.18653/v1/2022.gebnlp-1.3 · Cited by 11
Identifying and Mitigating Privacy Risks Stemming from Language Models: A Survey
Smith, V., Shamsabadi, A. S., Ashurst, C., & Weller, A. (2023) · arXiv · doi:10.48550/arxiv.2310.01424 · Cited by 10 · cited 2×
“Furthermore, the success of unlearning depends on the specific target data and the domain of the to-be-forgotten data (Smith et al 2023).”
Mind the GAP: A Balanced Corpus of Gendered Ambiguous PronounsPreprint
Webster, K., Recasens, M. V., Axelrod, V., & Baldridge, J. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1810.05201 · Cited by 9
Elastic Weight Removal for Faithful and Abstractive Dialogue Generation
Daheim, N., Dziri, N., Sachan, M., Gurevych, I., & Ponti, E. (2024) · Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers) · doi:10.18653/v1/2024.naacl-long.393 · Cited by 7
Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and TaxonomyPreprint
Shaik, T., Tao, X., Xie, H., et al. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2305.06360 · Cited by 6
Ni, S., Chen, D., Li, C., Hu, X., Xu, R., & Yang, M. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.310 · Cited by 6
Knowledge Unlearning for Mitigating Privacy Risks in Language Models
Jang, J., Yoon, D., Yang, S., Cha, S., Lee, M., Logeswaran, L., & Seo, M. (2022) · arXiv · doi:10.48550/arxiv.2210.01504 · Cited by 4
What can we learn from Data Leakage and Unlearning for Law?
Borkar, J. (2023) · arXiv · doi:10.48550/arxiv.2307.10476 · Cited by 4
Gradient Ascent Post-training Enhances Language Model Generalization
Yoon, D., Jang, J., Kim, S., & Seo, M. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/2023.acl-short.74 · Cited by 3
Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes
Gallegos, I. O., Rossi, R. A., Barrow, J., Tanjim, M. M., Yu, T., Deilamsalehy, H., Zhang, R., Kim, S., & Dernoncourt, F. (2024) · arXiv · doi:10.48550/arxiv.2402.01981 · Cited by 3
Limisiewicz, T., & Mareček, D. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2206.10744 · Cited by 2
Forgetting Private Textual Sequences in Language Models Via Leave-One-Out Ensemble
Liu, Z., & Kalinli, O. (2024) · ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp48485.2024.10446299 · Cited by 1 · cited 3×
“Liu and Kalinli (2023) propose the leave-one-out (LOO) ensemble method to reduce retraining cost and maintain utility at inference time.”
Forgetting Private Textual Sequences in Language Models via Leave-One-Out Ensemble
Liu, Z., & Kalinli, O. (2023) · arXiv · doi:10.48550/arxiv.2309.16082 · Cited by 1
Elastic Weight Removal for Faithful and Abstractive Dialogue Generation
Daheim, N., Dziri, N., Sachan, M., Gurevych, I., & Ponti, E. M. (2023) · arXiv · doi:10.48550/arxiv.2303.17574
Privacy Adhering Machine Un-learning in NLPPreprint
Kumar, V. B., Gangadharaiah, R., & Roth, D. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.09573
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
Workshop, B., :, Scao, T. L., Fan, A., Akiki, C., Pavlick, E., Ilić, S., Hesslow, D., Castagné, R., Luccioni, A. S., et al. (2022) · arXiv · doi:10.48550/arxiv.2211.05100
OPT: Open Pre-trained Transformer Language Models
Zhang, S., Roller, S., Goyal, N., Artetxe, M., Chen, M., Chen, S., Dewan, C., Diab, M., Li, X., Lin, X. V., et al. (2022) · arXiv · doi:10.48550/arxiv.2205.01068
SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization
Gliwa, B., Mochol, I., Biesek, M., et al. (2019) · arXiv · doi:10.48550/arxiv.1911.12237
DEPN: Detecting and Editing Privacy Neurons in Pretrained Language Models
Wu, X., Li, J., Xu, M., et al. (2023) · arXiv · doi:10.48550/arxiv.2310.20138
Gradient Ascent Post-training Enhances Language Model Generalization
Yoon, D., Jang, J., Kim, S., et al. (2023) · arXiv · doi:10.48550/arxiv.2306.07052
Self-Detoxifying Language Models via Toxification Reversal
Leong, C. T., Cheng, Y., Wang, J., et al. (2023) · arXiv · doi:10.48550/arxiv.2310.09573
Composing Parameter-Efficient Modules with Arithmetic Operation
Zhang, J., Chen, S., Liu, J., et al. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-0552
FairSISA: Ensemble Post-Processing to Improve Fairness of Unlearning in LLMs
Kadhe, S. R., Halimi, A., Rawat, A., & Baracaldo, N. (2023) · arXiv · doi:10.48550/arxiv.2312.07420
CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models
Nangia, N., Vania, C., Bhalerao, R., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.00133
RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models
Gehman, S., Gururangan, S., Sap, M., Choi, Y., & Smith, N. A. (2020) · arXiv · doi:10.48550/arxiv.2009.11462
Research works printed but not linked: 9
- Carlini N, Tramer F, Wallace E, et al (2021) Extracting training data from large language models. In: 30th USENIX Security Symposium (USENIX Security 21), pp 2633–2650
- Cettolo M, Niehues J, Stüker S, et al (2014) Report on the 11th IWSLT evaluation campaign. In: Federico M, Stüker S, Yvon F (eds) Proceedings of the 11th International Workshop on Spoken Language Translation: Evaluation Campaign@IWSLT 2014, Lake Tahoe, CA, USA, December 4-5, 2014, https://aclanthology.org/2014.iwslt-evaluation.1
- Commission European (2019) Ethics guidelines for trustworthy AI. Publications Office of the EU, doi/. https://doi.org/10.2759/346720 · doi:10.2759/346720
- Gordon A, Kozareva Z, Roemmele M (2012) SemEval-2012 task 7: Choice of plausible alternatives: An evaluation of commonsense causal reasoning. In: Agirre E, Bos J, Diab M, et al (eds) *SEM 2012: The First Joint Conference on Lexical and Computational Semantics – Volume 1: Proceedings of the main conference and the shared task, and Volume 2: Proceedings of the Sixth International Workshop on Semantic Evaluation (SemEval 2012). Association for Computational Linguistics, Montréal, Canada, pp 394–398, https://aclanthology.org/S12-1052
- Klimt B, Yang Y (2004) Introducing the enron corpus. In: CEAS 2004 - First Conference on Email and Anti-Spam, July 30-31, 2004, Mountain View, California, USA, http://www.ceas.cc/papers-2004/168.pdf
- Maas AL, Daly RE, Pham PT, et al (2011) Learning word vectors for sentiment analysis. In: Lin D, Matsumoto Y, Mihalcea R (eds) The 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference, 19-24 June, 2011, Portland, Oregon, USA. The Association for Computer Linguistics, pp 142–150, https://aclanthology.org/P11-1015/
- Radford A, Wu J, Child R et al (2019) Language models are unsupervised multitask learners. OpenAI blog 1(8):9
- Raffel C, Shazeer N, Roberts A et al (2020) Exploring the limits of transfer learning with a unified text-to-text transformer. J Mach Learn Res 21(1):5485–5551
- Rowling JK (2000) Harry Potter and the Sorcerer’s Stone. Bloomsbury
Cited by
A Survey of Machine Unlearning
Nguyen, T. T., Huynh, T. T., Ren, Z., Nguyen, P. L., Liew, A. W.-C., Yin, H., & Nguyen, Q. V. H. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3749987 · Cited by 164
A Survey of Machine Unlearning
Nguyen, T. T., Huynh, T. T., Ren, Z., Nguyen, P. L., Liew, A. W.-C., Yin, H., & Nguyen, Q. V. H. (2022) · arXiv · doi:10.48550/arxiv.2209.02299 · Cited by 193
Machine Unlearning: Taxonomy, Metrics, Applications, Challenges, and Prospects
Li, N., Zhou, C., Gao, Y., Chen, H., Zhang, Z., Kuang, B., & Fu, A. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2025.3530988 · Cited by 132
“Studies indicate that LLMs can generate biased output, such as “men are suited to be engineers, women to be nurses” [27].”
Road of Large Language Model: Source, Challenge, and Future Perspectives
Zhao, W., Yang, X., Lv, Z., Xu, C. X., & Guan, Z. (2025) · Research · doi:10.34133/research.0655 · Cited by 21
“Therefore, using methods such as unlearning [ 40 ] or contrastive learning [ 41 ], we can guide LLMs to avoid generating negative evaluation outputs, thereby achieving human preference alignment in complex scenarios.”
Position: LLM Unlearning Benchmarks are Weak Measures of Progress
Thaker, P., Hu, S., Kale, N., Maurya, Y., Wu, Z. S., & Smith, V. (2025) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml64287.2025.00035 · Cited by 78
“We point the reader to a number of comprehensive surveys on unlearning for further background, including [18, 19, 20, 21, 22, 23, 24, 25, 26, 27].”
Metadata
- Authors
- Blanco-Justicia, A., Jebreel, N., Manzanares-Salor, B., Sánchez, D., Domingo-Ferrer, J., Collell, G., & Eeik Tan, K.
- Affiliations
- Universitat Rovira i Virgili (ES)
- Corresponding author
- Sánchez, David
- Venue
- Artificial Intelligence Review
- Published
- 2025-01-13
- Volume
- 58
- Issue
- 3
- ISSN
- 0269-2821 (ISSN-L)
- 1573-7462
- OpenAlex
- W4406299883
- Semantic Scholar
- 41b4cd49570fc4330b600d85da1994b0532a8f95
- Cited by
- 31
- References
- 67
- Citations by year
- 2026: 16 · 2025: 15
- FWCI
- 12.54
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- Artificial Intelligence Review (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, other-oa)
- Rovira & Virgili repository (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- Universitat Rovira i Virgili
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- machine unlearning, large language models, AI regulation, AI risk mitigation
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.