Recent language Models (LMs) have shown impressive capabilities in generating texts with the knowledge internalized in parameters. Yet, LMs often generate the factually incorrect responses to the given queries, since ...
详细信息
ISBN:
(纸本)9798891760608
Recent language Models (LMs) have shown impressive capabilities in generating texts with the knowledge internalized in parameters. Yet, LMs often generate the factually incorrect responses to the given queries, since their knowledge may be inaccurate, incomplete, and outdated. To address this problem, previous works propose to augment LMs with the knowledge retrieved from an external knowledge source. However, such approaches often show suboptimal text generation performance due to two reasons: 1) the model may fail to retrieve the knowledge relevant to the given query, or 2) the model may not faithfully reflect the retrieved knowledge in the generated text. To overcome these, we propose to verify the output and the knowledge of the knowledge-augmented LMs with a separate verifier, which is a small LM that is trained to detect those two types of errors through instruction-finetuning. Then, when the verifier recognizes an error, we can rectify it by either retrieving new knowledge or generating new text. Further, we use an ensemble of the outputs from different instructions with a single verifier to enhance the reliability of the verification processes. We validate the effectiveness of the proposed verification steps on multiple question answering benchmarks, whose results show that the proposed verifier effectively identifies retrieval and generation errors, allowing LMs to provide more factually correct outputs. Our code is available at https://***/JinheonBaek/KALMV.
This work proposes a new naturallanguageprocessing (NLP) task to tackle the issue of multi-round, sequential text-based knowledge update. The study introduces a hybrid learning architecture and a novel self-supervis...
详细信息
ISBN:
(纸本)9798400701245
This work proposes a new naturallanguageprocessing (NLP) task to tackle the issue of multi-round, sequential text-based knowledge update. The study introduces a hybrid learning architecture and a novel self-supervised training strategy to enable generative language models to consolidate knowledge in the same way as humans. A dataset was also created for evaluation and results showed the effectiveness of our methodology. Experimental results confirm the superiority of the proposed approach over existing models and large language models (LLMs). The proposed task and model framework have the potential to significantly improve the automation of knowledge organization, making text-based knowledge an increasingly crucial resource for powerful LLMs to perform various tasks for humans.
Synthetic lethality (SL) offers a promising approach for targeted anti-cancer therapy. Deeply understanding SL gene pair mechanisms is vital for anti-cancer drug discovery. However, current wet-lab and machine learnin...
详细信息
Linguistic steganography is a key information hiding technique but faces challenges like abrupt content shifts, detection risks, and high training resource demands. To address these, this paper introduces SCF-Stega, a...
详细信息
Even though large language models (LLMs) have demonstrated remarkable performance across various naturallanguageprocessing tasks, their application in speech-related tasks has largely remained underexplored. This wo...
详细信息
Software engineering (SE) contracts play a pivotal role in Information Technology Outsourcing (ITO) projects. The obligations in SE contracts are known to be a useful source for deriving software requirements, thereby...
详细信息
ISBN:
(纸本)9798350395129;9798350395112
Software engineering (SE) contracts play a pivotal role in Information Technology Outsourcing (ITO) projects. The obligations in SE contracts are known to be a useful source for deriving software requirements, thereby contributing to the overall Software Development Life Cycle (SDLC). Making sense of contractual obligations is an important first step in successfully executing software projects. This includes building compliant systems, meeting delivery deadlines, avoiding heavy penalties, and steering clear of expensive litigations. In this work, we present an approach to capture the essence of a contractual clause by extracting its Contracts Grammar. Through an exploratory study, we first identify the constituents of Contracts Grammar. Subsequently, we experiment with multiple approaches for the automated extraction of these constituents, including extractive question-answering, token classification, text-to-text generation, prompting, and regular expressions. The question-answering based approach performed the best in terms of high average ROUGE-L score of 0.81, and faster inference times. The work presented in this paper is a part of the Contracts Governance System (CGS) and is in the process of deployment within a large IT vendor organization.
knowledge graphs represent a potent instrument for the classification and exhibition of data, as they encompass a systematic approach for the containment and retrieval of multifarious datasets. In finance, the utiliza...
详细信息
The rise of social networking services has increased text-based communication, often leading to misunderstandings. This study aims to develop a system using large language models (LLMs) like ChatGPT to provide real-ti...
详细信息
This research study presents a comprehensive survey of naturallanguageprocessing (NLP) research, tracing its historical evolution from its inception to the present. The survey explores the key milestones and advance...
详细信息
The proceedings contain 17 papers. The topics discussed include: on the relationship of social gender equality and grammatical gender in pre-trained large language models;the difficulty of misinformation labelling: a ...
The proceedings contain 17 papers. The topics discussed include: on the relationship of social gender equality and grammatical gender in pre-trained large language models;the difficulty of misinformation labelling: a case study for radon gas-related searches;findings of a machine translation shared task focused on Covid-19 related documents;COCOTEROS: a Spanish corpus with contextual knowledge for naturallanguage generation;emotions and news structure: an analysis of the language of fake news in Spanish;towards multi-class smishing detection: a novel feature vector approach and the Smishing-4C Dataset;synthetic annotated data for named entity recognition in computed tomography scan reports;Spanish FatPhoCorpus 2023: combating fatphobia in social media in Spanish using transformers;Spanish-language platform for drug-disease evidence search based on scientific articles;and automatic pathology detection in Spanish clinical notes combining language models and medical ontologies.
暂无评论