Related papers: Learning to Plan and Generate Text with Citations

Learning to Plan and Generate Text with Citations

URL: http://arxiv.org/abs/2404.03381v3
Date: Tue, 23 Jul 2024 11:54:10 GMT
Title: Learning to Plan and Generate Text with Citations
Authors: Constanza Fierro, Reinald Kim Amplayo, Fantine Huot, Nicola De Cao, Joshua Maynez, Shashi Narayan, Mirella Lapata,
Abstract summary: We explore the attribution capabilities of plan-based models which have been recently shown to improve the faithfulness, grounding, and controllability of generated text. We propose two attribution models that utilize different variants of blueprints, an abstractive model where questions are generated from scratch, and an extractive model where questions are copied from the input.
Score: 69.56850173097116
License: http://creativecommons.org/licenses/by/4.0/
Abstract: The increasing demand for the deployment of LLMs in information-seeking scenarios has spurred efforts in creating verifiable systems, which generate responses to queries along with supporting evidence. In this paper, we explore the attribution capabilities of plan-based models which have been recently shown to improve the faithfulness, grounding, and controllability of generated text. We conceptualize plans as a sequence of questions which serve as blueprints of the generated content and its organization. We propose two attribution models that utilize different variants of blueprints, an abstractive model where questions are generated from scratch, and an extractive model where questions are copied from the input. Experiments on long-form question-answering show that planning consistently improves attribution quality. Moreover, the citations generated by blueprint models are more accurate compared to those obtained from LLM-based pipelines lacking a planning component.

Related papers

Analysis of Plan-based Retrieval for Grounded Text Generation [78.89478272104739]
hallucinations occur when a language model is given a generation task outside its parametric knowledge. A common strategy to address this limitation is to infuse the language models with retrieval mechanisms. We analyze how planning can be used to guide retrieval to further reduce the frequency of hallucinations.
arXiv Detail & Related papers (2024-08-20T02:19:35Z)
PLANTS: A Novel Problem and Dataset for Summarization of Planning-Like (PL) Tasks [6.408228255254908]
We introduce a novel plan summarization problem, presenting a dataset, and providing a baseline method for generating PL summaries. We believe the novel problem and dataset can reinvigorate research in summarization, which some consider as a solved problem.
arXiv Detail & Related papers (2024-07-18T15:36:02Z)
Retrieve-Plan-Generation: An Iterative Planning and Answering Framework for Knowledge-Intensive LLM Generation [47.22520829950929]
We propose the Retrieve-Plan-Generation (RPG) framework for large language models (LLMs) RPG generates plan tokens to guide subsequent generation in the plan stage. In the answer stage, the model selects relevant fine-grained paragraphs based on the plan and uses them for further answer generation.
arXiv Detail & Related papers (2024-06-21T08:45:52Z)
Learning to Plan for Retrieval-Augmented Large Language Models from Knowledge Graphs [59.76268575344119]
We introduce a novel framework for enhancing large language models' (LLMs) planning capabilities by using planning data derived from knowledge graphs (KGs) LLMs fine-tuned with KG data have improved planning capabilities, better equipping them to handle complex QA tasks that involve retrieval.
arXiv Detail & Related papers (2024-06-20T13:07:38Z)
Model Generation with LLMs: From Requirements to UML Sequence Diagrams [9.114284818139069]
This paper investigates the capability of ChatGPT to generate a specific type of model, i.e., sequence diagrams, from NL requirements. We examine the sequence diagrams generated by ChatGPT for 28 requirements documents of various types and from different domains. Our results indicate that, although the models generally conform to the standard and exhibit a reasonable level of understandability, their completeness and correctness with respect to the specified requirements often present challenges.
arXiv Detail & Related papers (2024-04-09T15:07:25Z)
Conditional Generation with a Question-Answering Blueprint [84.95981645040281]
We advocate planning as a useful intermediate representation for rendering conditional generation less opaque and more grounded. We obtain blueprints automatically by exploiting state-of-the-art question generation technology. We develop Transformer-based models, each varying in how they incorporate the blueprint in the generated output.
arXiv Detail & Related papers (2022-07-01T13:10:19Z)
Data-to-text Generation with Variational Sequential Planning [74.3955521225497]
We consider the task of data-to-text generation, which aims to create textual output from non-linguistic input. We propose a neural model enhanced with a planning component responsible for organizing high-level information in a coherent and meaningful way. We infer latent plans sequentially with a structured variational model, while interleaving the steps of planning and generation.
arXiv Detail & Related papers (2022-02-28T13:17:59Z)
Text Modular Networks: Learning to Decompose Tasks in the Language of Existing Models [61.480085460269514]
We propose a framework for building interpretable systems that learn to solve complex tasks by decomposing them into simpler ones solvable by existing models. We use this framework to build ModularQA, a system that can answer multi-hop reasoning questions by decomposing them into sub-questions answerable by a neural factoid single-span QA model and a symbolic calculator.
arXiv Detail & Related papers (2020-09-01T23:45:42Z)

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