Towards Best Practices for Open Datasets for LLM Training

• URL: http://arxiv.org/abs/2501.08365v1
• Date: Tue, 14 Jan 2025 17:18:05 GMT
• Title: Towards Best Practices for Open Datasets for LLM Training
• Authors: Stefan Baack, Stella Biderman, Kasia Odrozek, Aviya Skowron, Ayah Bdeir, Jillian Bommarito, Jennifer Ding, Maximilian Gahntz, Paul Keller, Pierre-Carl Langlais, Greg Lindahl, Sebastian Majstorovic, Nik Marda, Guilherme Penedo, Maarten Van Segbroeck, Jennifer Wang, Leandro von Werra, Mitchell Baker, Julie Belião, Kasia Chmielinski, Marzieh Fadaee, Lisa Gutermuth, Hynek Kydlíček, Greg Leppert, EM Lewis-Jong, Solana Larsen, Shayne Longpre, Angela Oduor Lungati, Cullen Miller, Victor Miller, Max Ryabinin, Kathleen Siminyu, Andrew Strait, Mark Surman, Anna Tumadóttir, Maurice Weber, Rebecca Weiss, Lee White, Thomas Wolf,
• Abstract summary: Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. Creative producers have led to several high-profile copyright lawsuits. This trend in limiting data information causes harm by hindering transparency, accountability, and innovation.
• Score: 21.448011162803866
• License:
• Abstract: Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. The permissibility of doing so varies by jurisdiction: in countries like the EU and Japan, this is allowed under certain restrictions, while in the United States, the legal landscape is more ambiguous. Regardless of the legal status, concerns from creative producers have led to several high-profile copyright lawsuits, and the threat of litigation is commonly cited as a reason for the recent trend towards minimizing the information shared about training datasets by both corporate and public interest actors. This trend in limiting data information causes harm by hindering transparency, accountability, and innovation in the broader ecosystem by denying researchers, auditors, and impacted individuals access to the information needed to understand AI models. While this could be mitigated by training language models on open access and public domain data, at the time of writing, there are no such models (trained at a meaningful scale) due to the substantial technical and sociological challenges in assembling the necessary corpus. These challenges include incomplete and unreliable metadata, the cost and complexity of digitizing physical records, and the diverse set of legal and technical skills required to ensure relevance and responsibility in a quickly changing landscape. Building towards a future where AI systems can be trained on openly licensed data that is responsibly curated and governed requires collaboration across legal, technical, and policy domains, along with investments in metadata standards, digitization, and fostering a culture of openness.

Related papers

• Towards Data Governance of Frontier AI Models [0.0]
  We look at how data can enable new governance capacities for frontier AI models. Data is non-rival, often non-excludable, easily replicable, and increasingly synthesizable. We propose a set of policy mechanisms targeting key actors along the data supply chain.
  arXiv  Detail & Related papers  (2024-12-05T02:37:51Z)
• Assessing Privacy Policies with AI: Ethical, Legal, and Technical Challenges [6.916147439085307]
  Large Language Models (LLMs) can be used to assess privacy policies for users automatically. We explore the challenges of this approach in three pillars, namely technical feasibility, ethical implications, and legal compatibility. Our findings aim to identify potential for future research, and to foster a discussion on the use of LLM technologies.
  arXiv  Detail & Related papers  (2024-10-10T21:36:35Z)
• The Future of International Data Transfers: Managing Legal Risk with a User-Held Data Model [0.0]
  The General Data Protection Regulation contains a blanket prohibition on the transfer of personal data outside of the European Economic Area unless strict requirements are met. New technologies have made international data transfers the norm and not the exception. This article examines one such alternative, namely a user-held data model.
  arXiv  Detail & Related papers  (2024-07-30T03:15:36Z)
• Unlearning Targeted Information via Single Layer Unlearning Gradient [15.374381635334897]
  Unauthorized privacy-related computation is a significant concern for society. The EU's General Protection Regulation includes a "right to be forgotten" We propose Single Layer Unlearning Gradient (SLUG) to unlearn targeted information by updating targeted layers of a model.
  arXiv  Detail & Related papers  (2024-07-16T15:52:36Z)
• InternLM-Law: An Open Source Chinese Legal Large Language Model [72.2589401309848]
  InternLM-Law is a specialized LLM tailored for addressing diverse legal queries related to Chinese laws. We meticulously construct a dataset in the Chinese legal domain, encompassing over 1 million queries. InternLM-Law achieves the highest average performance on LawBench, outperforming state-of-the-art models, including GPT-4, on 13 out of 20 subtasks.
  arXiv  Detail & Related papers  (2024-06-21T06:19:03Z)
• The Frontier of Data Erasure: Machine Unlearning for Large Language Models [56.26002631481726]
  Large Language Models (LLMs) are foundational to AI advancements. LLMs pose risks by potentially memorizing and disseminating sensitive, biased, or copyrighted information. Machine unlearning emerges as a cutting-edge solution to mitigate these concerns.
  arXiv  Detail & Related papers  (2024-03-23T09:26:15Z)
• The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI [41.32981860191232]
  Legal and machine learning experts to systematically audit and trace 1800+ text datasets. Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets. frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+.
  arXiv  Detail & Related papers  (2023-10-25T17:20:26Z)
• SILO Language Models: Isolating Legal Risk In a Nonparametric Datastore [159.21914121143885]
  We present SILO, a new language model that manages this risk-performance tradeoff during inference. SILO is built by (1) training a parametric LM on Open License Corpus (OLC), a new corpus we curate with 228B tokens of public domain and permissively licensed text. Access to the datastore greatly improves out of domain performance, closing 90% of the performance gap with an LM trained on the Pile.
  arXiv  Detail & Related papers  (2023-08-08T17:58:15Z)
• Distributed Machine Learning and the Semblance of Trust [66.1227776348216]
  Federated Learning (FL) allows the data owner to maintain data governance and perform model training locally without having to share their data. FL and related techniques are often described as privacy-preserving. We explain why this term is not appropriate and outline the risks associated with over-reliance on protocols that were not designed with formal definitions of privacy in mind.
  arXiv  Detail & Related papers  (2021-12-21T08:44:05Z)
• Representative & Fair Synthetic Data [68.8204255655161]
  We present a framework to incorporate fairness constraints into the self-supervised learning process. We generate a representative as well as fair version of the UCI Adult census data set. We consider representative & fair synthetic data a promising future building block to teach algorithms not on historic worlds, but rather on the worlds that we strive to live in.
  arXiv  Detail & Related papers  (2021-04-07T09:19:46Z)
• A vision for global privacy bridges: Technical and legal measures for international data markets [77.34726150561087]
  Despite data protection laws and an acknowledged right to privacy, trading personal information has become a business equated with "trading oil" An open conflict is arising between business demands for data and a desire for privacy. We propose and test a vision of a personal information market with privacy.
  arXiv  Detail & Related papers  (2020-05-13T13:55:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.