Pub Date : 2023-08-16DOI: 10.18653/v1/2023.bionlp-1.31
Vera Pavlova, M. Makhlouf
Using language models (LMs) pre-trained in a self-supervised setting on large corpora and then fine-tuning for a downstream task has helped to deal with the problem of limited label data for supervised learning tasks such as Named Entity Recognition (NER). Recent research in biomedical language processing has offered a number of biomedical LMs pre-trained using different methods and techniques that advance results on many BioNLP tasks, including NER. However, there is still a lack of a comprehensive comparison of pre-training approaches that would work more optimally in the biomedical domain. This paper aims to investigate different pre-training methods, such as pre-training the biomedical LM from scratch and pre-training it in a continued fashion. We compare existing methods with our proposed pre-training method of initializing weights for new tokens by distilling existing weights from the BERT model inside the context where the tokens were found. The method helps to speed up the pre-training stage and improve performance on NER. In addition, we compare how masking rate, corruption strategy, and masking strategies impact the performance of the biomedical LM. Finally, using the insights from our experiments, we introduce a new biomedical LM (BIOptimus), which is pre-trained using Curriculum Learning (CL) and contextualized weight distillation method. Our model sets new states of the art on several biomedical Named Entity Recognition (NER) tasks. We release our code and all pre-trained models.
{"title":"BIOptimus: Pre-training an Optimal Biomedical Language Model with Curriculum Learning for Named Entity Recognition","authors":"Vera Pavlova, M. Makhlouf","doi":"10.18653/v1/2023.bionlp-1.31","DOIUrl":"https://doi.org/10.18653/v1/2023.bionlp-1.31","url":null,"abstract":"Using language models (LMs) pre-trained in a self-supervised setting on large corpora and then fine-tuning for a downstream task has helped to deal with the problem of limited label data for supervised learning tasks such as Named Entity Recognition (NER). Recent research in biomedical language processing has offered a number of biomedical LMs pre-trained using different methods and techniques that advance results on many BioNLP tasks, including NER. However, there is still a lack of a comprehensive comparison of pre-training approaches that would work more optimally in the biomedical domain. This paper aims to investigate different pre-training methods, such as pre-training the biomedical LM from scratch and pre-training it in a continued fashion. We compare existing methods with our proposed pre-training method of initializing weights for new tokens by distilling existing weights from the BERT model inside the context where the tokens were found. The method helps to speed up the pre-training stage and improve performance on NER. In addition, we compare how masking rate, corruption strategy, and masking strategies impact the performance of the biomedical LM. Finally, using the insights from our experiments, we introduce a new biomedical LM (BIOptimus), which is pre-trained using Curriculum Learning (CL) and contextualized weight distillation method. Our model sets new states of the art on several biomedical Named Entity Recognition (NER) tasks. We release our code and all pre-trained models.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114576888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-05DOI: 10.18653/v1/2023.bionlp-1.1
Yueling Li, Sebastian Martschat, Simone Paolo Ponzetto
We present a cross-domain approach for automated measurement and context extraction based on pre-trained language models. We construct a multi-source, multi-domain corpus and train an end-to-end extraction pipeline. We then apply multi-source task-adaptive pre-training and fine-tuning to benchmark the cross-domain generalization capability of our model. Further, we conceptualize and apply a task-specific error analysis and derive insights for future work. Our results suggest that multi-source training leads to the best overall results, while single-source training yields the best results for the respective individual domain. While our setup is successful at extracting quantity values and units, more research is needed to improve the extraction of contextual entities. We make the cross-domain corpus used in this work available online.
{"title":"Multi-Source (Pre-)Training for Cross-Domain Measurement, Unit and Context Extraction","authors":"Yueling Li, Sebastian Martschat, Simone Paolo Ponzetto","doi":"10.18653/v1/2023.bionlp-1.1","DOIUrl":"https://doi.org/10.18653/v1/2023.bionlp-1.1","url":null,"abstract":"We present a cross-domain approach for automated measurement and context extraction based on pre-trained language models. We construct a multi-source, multi-domain corpus and train an end-to-end extraction pipeline. We then apply multi-source task-adaptive pre-training and fine-tuning to benchmark the cross-domain generalization capability of our model. Further, we conceptualize and apply a task-specific error analysis and derive insights for future work. Our results suggest that multi-source training leads to the best overall results, while single-source training yields the best results for the respective individual domain. While our setup is successful at extracting quantity values and units, more research is needed to improve the extraction of contextual entities. We make the cross-domain corpus used in this work available online.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130479245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-14DOI: 10.48550/arXiv.2306.08403
Oumaima El Khettari, Solen Quiniou, Samuel Chaffron
We present a manually annotated new corpus, Species-Species Interaction (SSI), for extracting meaningful binary relations between species, in biomedical texts, at sentence level, with a focus on the gut microbiota. The corpus leverages PubTator to annotate species in full-text articles after evaluating different NER species taggers. Our first results are promising for extracting relations between species using BERT and its biomedical variants.
{"title":"Building a Corpus for Biomedical Relation Extraction of Species Mentions","authors":"Oumaima El Khettari, Solen Quiniou, Samuel Chaffron","doi":"10.48550/arXiv.2306.08403","DOIUrl":"https://doi.org/10.48550/arXiv.2306.08403","url":null,"abstract":"We present a manually annotated new corpus, Species-Species Interaction (SSI), for extracting meaningful binary relations between species, in biomedical texts, at sentence level, with a focus on the gut microbiota. The corpus leverages PubTator to annotate species in full-text articles after evaluating different NER species taggers. Our first results are promising for extracting relations between species using BERT and its biomedical variants.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115217013","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-07DOI: 10.48550/arXiv.2306.04820
S. Chandrasekhar, Chieh-Yang Huang, Ting Huang
The rapid growth of scientific publications, particularly during the COVID-19 pandemic, emphasizes the need for tools to help researchers efficiently comprehend the latest advancements. One essential part of understanding scientific literature is research aspect classification, which categorizes sentences in abstracts to Background, Purpose, Method, and Finding. In this study, we investigate the impact of different datasets on model performance for the crowd-annotated CODA-19 research aspect classification task. Specifically, we explore the potential benefits of using the large, automatically curated PubMed 200K RCT dataset and evaluate the effectiveness of large language models (LLMs), such as LLaMA, GPT-3, ChatGPT, and GPT-4. Our results indicate that using the PubMed 200K RCT dataset does not improve performance for the CODA-19 task. We also observe that while GPT-4 performs well, it does not outperform the SciBERT model fine-tuned on the CODA-19 dataset, emphasizing the importance of a dedicated and task-aligned datasets dataset for the target task.
{"title":"Good Data, Large Data, or No Data? Comparing Three Approaches in Developing Research Aspect Classifiers for Biomedical Papers","authors":"S. Chandrasekhar, Chieh-Yang Huang, Ting Huang","doi":"10.48550/arXiv.2306.04820","DOIUrl":"https://doi.org/10.48550/arXiv.2306.04820","url":null,"abstract":"The rapid growth of scientific publications, particularly during the COVID-19 pandemic, emphasizes the need for tools to help researchers efficiently comprehend the latest advancements. One essential part of understanding scientific literature is research aspect classification, which categorizes sentences in abstracts to Background, Purpose, Method, and Finding. In this study, we investigate the impact of different datasets on model performance for the crowd-annotated CODA-19 research aspect classification task. Specifically, we explore the potential benefits of using the large, automatically curated PubMed 200K RCT dataset and evaluate the effectiveness of large language models (LLMs), such as LLaMA, GPT-3, ChatGPT, and GPT-4. Our results indicate that using the PubMed 200K RCT dataset does not improve performance for the CODA-19 task. We also observe that while GPT-4 performs well, it does not outperform the SciBERT model fine-tuned on the CODA-19 dataset, emphasizing the importance of a dedicated and task-aligned datasets dataset for the target task.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122908184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-07DOI: 10.48550/arXiv.2306.04504
Israt Jahan, Md Tahmid Rahman Laskar, Chun Peng, J. Huang
ChatGPT is a large language model developed by OpenAI. Despite its impressive performance across various tasks, no prior work has investigated its capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of ChatGPT on various benchmark biomedical tasks, such as relation extraction, document classification, question answering, and summarization. To the best of our knowledge, this is the first work that conducts an extensive evaluation of ChatGPT in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot ChatGPT even outperforms the state-of-the-art fine-tuned generative transformer models, such as BioGPT and BioBART. This suggests that ChatGPT’s pre-training on large text corpora makes it quite specialized even in the biomedical domain. Our findings demonstrate that ChatGPT has the potential to be a valuable tool for various tasks in the biomedical domain that lack large annotated data.
{"title":"Evaluation of ChatGPT on Biomedical Tasks: A Zero-Shot Comparison with Fine-Tuned Generative Transformers","authors":"Israt Jahan, Md Tahmid Rahman Laskar, Chun Peng, J. Huang","doi":"10.48550/arXiv.2306.04504","DOIUrl":"https://doi.org/10.48550/arXiv.2306.04504","url":null,"abstract":"ChatGPT is a large language model developed by OpenAI. Despite its impressive performance across various tasks, no prior work has investigated its capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of ChatGPT on various benchmark biomedical tasks, such as relation extraction, document classification, question answering, and summarization. To the best of our knowledge, this is the first work that conducts an extensive evaluation of ChatGPT in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot ChatGPT even outperforms the state-of-the-art fine-tuned generative transformer models, such as BioGPT and BioBART. This suggests that ChatGPT’s pre-training on large text corpora makes it quite specialized even in the biomedical domain. Our findings demonstrate that ChatGPT has the potential to be a valuable tool for various tasks in the biomedical domain that lack large annotated data.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125638692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-05DOI: 10.48550/arXiv.2306.03264
Sanjeev Kumar Karn, Rikhiya Ghosh, P. Kusuma, Oladimeji Farri
Instruction-tuned generative large language models (LLMs), such as ChatGPT and Bloomz, possess excellent generalization abilities. However, they face limitations in understanding radiology reports, particularly when generating the IMPRESSIONS section from the FINDINGS section. These models tend to produce either verbose or incomplete IMPRESSIONS, mainly due to insufficient exposure to medical text data during training. We present a system that leverages large-scale medical text data for domain-adaptive pre-training of instruction-tuned LLMs, enhancing their medical knowledge and performance on specific medical tasks. We demonstrate that this system performs better in a zero-shot setting compared to several pretrain-and-finetune adaptation methods on the IMPRESSIONS generation task. Furthermore, it ranks 1st among participating systems in Task 1B: Radiology Report Summarization.
{"title":"shs-nlp at RadSum23: Domain-Adaptive Pre-training of Instruction-tuned LLMs for Radiology Report Impression Generation","authors":"Sanjeev Kumar Karn, Rikhiya Ghosh, P. Kusuma, Oladimeji Farri","doi":"10.48550/arXiv.2306.03264","DOIUrl":"https://doi.org/10.48550/arXiv.2306.03264","url":null,"abstract":"Instruction-tuned generative large language models (LLMs), such as ChatGPT and Bloomz, possess excellent generalization abilities. However, they face limitations in understanding radiology reports, particularly when generating the IMPRESSIONS section from the FINDINGS section. These models tend to produce either verbose or incomplete IMPRESSIONS, mainly due to insufficient exposure to medical text data during training. We present a system that leverages large-scale medical text data for domain-adaptive pre-training of instruction-tuned LLMs, enhancing their medical knowledge and performance on specific medical tasks. We demonstrate that this system performs better in a zero-shot setting compared to several pretrain-and-finetune adaptation methods on the IMPRESSIONS generation task. Furthermore, it ranks 1st among participating systems in Task 1B: Radiology Report Summarization.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129839858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-05DOI: 10.48550/arXiv.2306.02754
Hao Li, Yuping Wu, Viktor Schlegel, R. Batista-Navarro, Thanh-Tung Nguyen, Abhinav Ramesh Kashyap, Xiaojun Zeng, Daniel Beck, Stefan Winkler, G. Nenadic
Medical progress notes play a crucial role in documenting a patient’s hospital journey, including his or her condition, treatment plan, and any updates for healthcare providers. Automatic summarisation of a patient’s problems in the form of a “problem list” can aid stakeholders in understanding a patient’s condition, reducing workload and cognitive bias. BioNLP 2023 Shared Task 1A focusses on generating a list of diagnoses and problems from the provider’s progress notes during hospitalisation. In this paper, we introduce our proposed approach to this task, which integrates two complementary components. One component employs large language models (LLMs) for data augmentation; the other is an abstractive summarisation LLM with a novel pre-training objective for generating the patients’ problems summarised as a list. Our approach was ranked second among all submissions to the shared task. The performance of our model on the development and test datasets shows that our approach is more robust on unknown data, with an improvement of up to 3.1 points over the same size of the larger model.
{"title":"Team:PULSAR at ProbSum 2023:PULSAR: Pre-training with Extracted Healthcare Terms for Summarising Patients’ Problems and Data Augmentation with Black-box Large Language Models","authors":"Hao Li, Yuping Wu, Viktor Schlegel, R. Batista-Navarro, Thanh-Tung Nguyen, Abhinav Ramesh Kashyap, Xiaojun Zeng, Daniel Beck, Stefan Winkler, G. Nenadic","doi":"10.48550/arXiv.2306.02754","DOIUrl":"https://doi.org/10.48550/arXiv.2306.02754","url":null,"abstract":"Medical progress notes play a crucial role in documenting a patient’s hospital journey, including his or her condition, treatment plan, and any updates for healthcare providers. Automatic summarisation of a patient’s problems in the form of a “problem list” can aid stakeholders in understanding a patient’s condition, reducing workload and cognitive bias. BioNLP 2023 Shared Task 1A focusses on generating a list of diagnoses and problems from the provider’s progress notes during hospitalisation. In this paper, we introduce our proposed approach to this task, which integrates two complementary components. One component employs large language models (LLMs) for data augmentation; the other is an abstractive summarisation LLM with a novel pre-training objective for generating the patients’ problems summarised as a list. Our approach was ranked second among all submissions to the shared task. The performance of our model on the development and test datasets shows that our approach is more robust on unknown data, with an improvement of up to 3.1 points over the same size of the larger model.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131104235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.48550/arXiv.2306.00665
François Remy, Thomas Demeester
Background: More than 400.000 biomedical concepts and some of their relationships are contained in SnomedCT, a comprehensive biomedical ontology. However, their concept names are not always readily interpretable by non-experts, or patients looking at their own electronic health records (EHR). Clear definitions or descriptions in understandable language or often not available. Therefore, generating human-readable definitions for biomedical concepts might help make the information they encode more accessible and understandable to a wider public.Objective: In this article, we introduce the Automatic Glossary of Clinical Terminology (AGCT), a large-scale biomedical dictionary of clinical concepts generated using high-quality information extracted from the biomedical knowledge contained in SnomedCT.Methods: We generate a novel definition for every SnomedCT concept, after prompting the OpenAI Turbo model, a variant of GPT 3.5, using a high-quality verbalization of the SnomedCT relationships of the to-be-defined concept. A significant subset of the generated definitions was subsequently evaluated by NLP researchers with biomedical expertise on 5-point scales along the following three axes: factuality, insight, and fluency.Results: AGCT contains 422,070 computer-generated definitions for SnomedCT concepts, covering various domains such as diseases, procedures, drugs, and anatomy. The average length of the definitions is 49 words. The definitions were assigned average scores of over 4.5 out of 5 on all three axes, indicating a majority of factual, insightful, and fluent definitions.Conclusion: AGCT is a novel and valuable resource for biomedical tasks that require human-readable definitions for SnomedCT concepts. It can also serve as a base for developing robust biomedical retrieval models or other applications that leverage natural language understanding of biomedical knowledge.
{"title":"Automatic Glossary of Clinical Terminology: a Large-Scale Dictionary of Biomedical Definitions Generated from Ontological Knowledge","authors":"François Remy, Thomas Demeester","doi":"10.48550/arXiv.2306.00665","DOIUrl":"https://doi.org/10.48550/arXiv.2306.00665","url":null,"abstract":"Background: More than 400.000 biomedical concepts and some of their relationships are contained in SnomedCT, a comprehensive biomedical ontology. However, their concept names are not always readily interpretable by non-experts, or patients looking at their own electronic health records (EHR). Clear definitions or descriptions in understandable language or often not available. Therefore, generating human-readable definitions for biomedical concepts might help make the information they encode more accessible and understandable to a wider public.Objective: In this article, we introduce the Automatic Glossary of Clinical Terminology (AGCT), a large-scale biomedical dictionary of clinical concepts generated using high-quality information extracted from the biomedical knowledge contained in SnomedCT.Methods: We generate a novel definition for every SnomedCT concept, after prompting the OpenAI Turbo model, a variant of GPT 3.5, using a high-quality verbalization of the SnomedCT relationships of the to-be-defined concept. A significant subset of the generated definitions was subsequently evaluated by NLP researchers with biomedical expertise on 5-point scales along the following three axes: factuality, insight, and fluency.Results: AGCT contains 422,070 computer-generated definitions for SnomedCT concepts, covering various domains such as diseases, procedures, drugs, and anatomy. The average length of the definitions is 49 words. The definitions were assigned average scores of over 4.5 out of 5 on all three axes, indicating a majority of factual, insightful, and fluent definitions.Conclusion: AGCT is a novel and valuable resource for biomedical tasks that require human-readable definitions for SnomedCT concepts. It can also serve as a base for developing robust biomedical retrieval models or other applications that leverage natural language understanding of biomedical knowledge.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116537898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-30DOI: 10.48550/arXiv.2305.19120
Harsh Verma, S. Bergler, Narjes Tahaei
We compare three simple and popular approaches for NER: 1) SEQ (sequence labeling with a linear token classifier) 2) SeqCRF (sequence labeling with Conditional Random Fields), and 3) SpanPred (span prediction with boundary token embeddings). We compare the approaches on 4 biomedical NER tasks: GENIA, NCBI-Disease, LivingNER (Spanish), and SocialDisNER (Spanish). The SpanPred model demonstrates state-of-the-art performance on LivingNER and SocialDisNER, improving F1 by 1.3 and 0.6 F1 respectively. The SeqCRF model also demonstrates state-of-the-art performance on LivingNER and SocialDisNER, improving F1 by 0.2 F1 and 0.7 respectively. The SEQ model is competitive with the state-of-the-art on LivingNER dataset. We explore some simple ways of combining the three approaches. We find that majority voting consistently gives high precision and high F1 across all 4 datasets.Lastly, we implement a system that learns to combine SEQ’s and SpanPred’s predictions, generating systems that give high recall and high F1 across all 4 datasets. On the GENIA dataset, we find that our learned combiner system significantly boosts F1(+1.2) and recall(+2.1) over the systems being combined.
{"title":"Comparing and combining some popular NER approaches on Biomedical tasks","authors":"Harsh Verma, S. Bergler, Narjes Tahaei","doi":"10.48550/arXiv.2305.19120","DOIUrl":"https://doi.org/10.48550/arXiv.2305.19120","url":null,"abstract":"We compare three simple and popular approaches for NER: 1) SEQ (sequence labeling with a linear token classifier) 2) SeqCRF (sequence labeling with Conditional Random Fields), and 3) SpanPred (span prediction with boundary token embeddings). We compare the approaches on 4 biomedical NER tasks: GENIA, NCBI-Disease, LivingNER (Spanish), and SocialDisNER (Spanish). The SpanPred model demonstrates state-of-the-art performance on LivingNER and SocialDisNER, improving F1 by 1.3 and 0.6 F1 respectively. The SeqCRF model also demonstrates state-of-the-art performance on LivingNER and SocialDisNER, improving F1 by 0.2 F1 and 0.7 respectively. The SEQ model is competitive with the state-of-the-art on LivingNER dataset. We explore some simple ways of combining the three approaches. We find that majority voting consistently gives high precision and high F1 across all 4 datasets.Lastly, we implement a system that learns to combine SEQ’s and SpanPred’s predictions, generating systems that give high recall and high F1 across all 4 datasets. On the GENIA dataset, we find that our learned combiner system significantly boosts F1(+1.2) and recall(+2.1) over the systems being combined.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133666308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-11DOI: 10.48550/arXiv.2304.05454
Chenhan Yuan, Qianqian Xie, S. Ananiadou
The goal of temporal relation extraction is to infer the temporal relation between two events in the document. Supervised models are dominant in this task. In this work, we investigate ChatGPT’s ability on zero-shot temporal relation extraction. We designed three different prompt techniques to break down the task and evaluate ChatGPT. Our experiments show that ChatGPT’s performance has a large gap with that of supervised methods and can heavily rely on the design of prompts. We further demonstrate that ChatGPT can infer more small relation classes correctly than supervised methods. The current shortcomings of ChatGPT on temporal relation extraction are also discussed in this paper. We found that ChatGPT cannot keep consistency during temporal inference and it fails in actively long-dependency temporal inference.
{"title":"Zero-shot Temporal Relation Extraction with ChatGPT","authors":"Chenhan Yuan, Qianqian Xie, S. Ananiadou","doi":"10.48550/arXiv.2304.05454","DOIUrl":"https://doi.org/10.48550/arXiv.2304.05454","url":null,"abstract":"The goal of temporal relation extraction is to infer the temporal relation between two events in the document. Supervised models are dominant in this task. In this work, we investigate ChatGPT’s ability on zero-shot temporal relation extraction. We designed three different prompt techniques to break down the task and evaluate ChatGPT. Our experiments show that ChatGPT’s performance has a large gap with that of supervised methods and can heavily rely on the design of prompts. We further demonstrate that ChatGPT can infer more small relation classes correctly than supervised methods. The current shortcomings of ChatGPT on temporal relation extraction are also discussed in this paper. We found that ChatGPT cannot keep consistency during temporal inference and it fails in actively long-dependency temporal inference.","PeriodicalId":200974,"journal":{"name":"Workshop on Biomedical Natural Language Processing","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127060032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：