Dual use issues in the field of Natural Language Generation

• URL: http://arxiv.org/abs/2501.06636v1
• Date: Sat, 11 Jan 2025 20:55:35 GMT
• Title: Dual use issues in the field of Natural Language Generation
• Authors: Emiel van Miltenburg,
• Abstract summary: This report documents the results of a recent survey in the SIGGEN community, focusing on Dual Use issues in Natural Language Generation (NLG)<n>With 23 respondents, the survey is presumably not representative of all SIGGEN members, but at least this document offers a helpful resource for future discussions.
• Score: 2.1427900765416985
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This report documents the results of a recent survey in the SIGGEN community, focusing on Dual Use issues in Natural Language Generation (NLG). SIGGEN is the Special Interest Group (SIG) of the Association for Computational Linguistics (ACL) for researchers working on NLG. The survey was prompted by the ACL executive board, which asked all SIGs to provide an overview of dual use issues within their respective subfields. The survey was sent out in October 2024 and the results were processed in January 2025. With 23 respondents, the survey is presumably not representative of all SIGGEN members, but at least this document offers a helpful resource for future discussions. This report is open to feedback from the SIGGEN community. Let me know if you have any questions or comments!

Related papers

• Retrieval-Augmented Generation with Conflicting Evidence [57.66282463340297]
  Large language model (LLM) agents are increasingly employing retrieval-augmented generation (RAG) to improve the factuality of their responses. In practice, these systems often need to handle ambiguous user queries and potentially conflicting information from multiple sources. We propose RAMDocs (Retrieval with Ambiguity and Misinformation in Documents), a new dataset that simulates complex and realistic scenarios for conflicting evidence for a user query.
  arXiv  Detail & Related papers  (2025-04-17T16:46:11Z)
• LM Agents for Coordinating Multi-User Information Gathering [82.3543678605684]
  PeopleJoin is a benchmark for evaluating LM-mediated collaborative problem solving. PeopleJoin comprises two evaluation domains: PeopleJoin-QA and PeopleJoin-DocCreation.
  arXiv  Detail & Related papers  (2025-02-17T21:19:45Z)
• Uchaguzi-2022: A Dataset of Citizen Reports on the 2022 Kenyan Election [49.35115948941981]
  We present Uchaguzi-2022, a dataset of 14k categorized and geotagged citizen reports related to the 2022 Kenyan General Election.<n>We use this dataset to investigate whether language models can assist in scalably categorizing and geotagging reports, thus highlighting its potential application in the AI for Social Good space.
  arXiv  Detail & Related papers  (2024-12-17T17:08:35Z)
• Usefulness of LLMs as an Author Checklist Assistant for Scientific Papers: NeurIPS'24 Experiment [59.09144776166979]
  Large language models (LLMs) represent a promising, but controversial, tool in aiding scientific peer review. This study evaluates the usefulness of LLMs in a conference setting as a tool for vetting paper submissions against submission standards.
  arXiv  Detail & Related papers  (2024-11-05T18:58:00Z)
• TREC iKAT 2023: The Interactive Knowledge Assistance Track Overview [11.276981461219515]
  iKAT emphasizes the creation and research of conversational search agents that adapt responses based on the user's prior interactions and present context. Most of the runs leveraged Large Language Models (LLMs) in their pipelines, with a few focusing on a generate-then-retrieve approach.
  arXiv  Detail & Related papers  (2024-01-02T18:40:03Z)
• SOUL: Towards Sentiment and Opinion Understanding of Language [96.74878032417054]
  We propose a new task called Sentiment and Opinion Understanding of Language (SOUL) SOUL aims to evaluate sentiment understanding through two subtasks: Review (RC) and Justification Generation (JG)
  arXiv  Detail & Related papers  (2023-10-27T06:48:48Z)
• The Shifted and The Overlooked: A Task-oriented Investigation of User-GPT Interactions [114.67699010359637]
  We analyze a large-scale collection of real user queries to GPT. We find that tasks such as design'' and planning'' are prevalent in user interactions but are largely neglected or different from traditional NLP benchmarks.
  arXiv  Detail & Related papers  (2023-10-19T02:12:17Z)
• Beyond Good Intentions: Reporting the Research Landscape of NLP for Social Good [115.1507728564964]
  We introduce NLP4SG Papers, a scientific dataset with three associated tasks. These tasks help identify NLP4SG papers and characterize the NLP4SG landscape. We use state-of-the-art NLP models to address each of these tasks and use them on the entire ACL Anthology.
  arXiv  Detail & Related papers  (2023-05-09T14:16:25Z)
• The impact and applications of ChatGPT: a systematic review of literature reviews [0.0]
  ChatGPT has become one of the most widely used natural language processing tools. With thousands of published papers demonstrating its applications across various industries and fields, ChatGPT has sparked significant interest in the research community. An overview of the available evidence from multiple reviews and studies could provide further insights, minimize redundancy, and identify areas where further research is needed.
  arXiv  Detail & Related papers  (2023-05-08T17:57:34Z)
• What Do NLP Researchers Believe? Results of the NLP Community Metasurvey [43.763865378178245]
  We present the results of the NLP Community Metasurvey. The survey elicited opinions on controversial issues. We find false sociological beliefs where the community's predictions don't match reality.
  arXiv  Detail & Related papers  (2022-08-26T19:45:51Z)
• Overview of ADoBo 2021: Automatic Detection of Unassimilated Borrowings in the Spanish Press [8.950918531231158]
  This paper summarizes the main findings of the ADoBo 2021 shared task, proposed in the context of IberLef 2021. In this task, we invited participants to detect lexical borrowings (coming mostly from English) in Spanish newswire texts. We provided participants with an annotated corpus of lexical borrowings which we split into training, development and test splits.
  arXiv  Detail & Related papers  (2021-10-29T11:07:59Z)

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.