Reading Comprehension
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 20
- 평가표
- 7
- 보존된 평가 기록
- 91
Most current question answering datasets frame the task as reading comprehension where the question is about a paragraph or document and the answer often is a span in the document.
Some specific tasks of reading comprehension include multi-modal machine reading comprehension and textual machine reading comprehension, among others. In the literature, machine reading comprehension can be divide into four categories: cloze style, multiple choice, span prediction, and free-form answer. Read more about each category here.
Benchmark datasets used for testing a model's reading comprehension abilities include MovieQA, ReCoRD, and RACE, among others.
The Machine Reading group at UCL also provides an overview of reading comprehension tasks.
Figure source: A Survey on Machine Reading Comprehension: Tasks, Evaluation Metrics and Benchmark Datasets
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
A Coordination-based Approach for Focused Learning in Knowledge-Based Systems
GoBERT: Gene Ontology Graph Informed BERT for Universal Gene Function Prediction
CarbonChat: Large Language Model-Based Corporate Carbon Emission Analysis and Climate Knowledge Q&A System
"My life is miserable, have to sign 500 autographs everyday": Exposing Humblebragging, the Brags in Disguise
Unlocking the Potential of Multiple BERT Models for Bangla Question Answering in NCTB Textbooks
Why We Build Local Large Language Models: An Observational Analysis from 35 Japanese and Multilingual LLMs
SKETCH: Structured Knowledge Enhanced Text Comprehension for Holistic Retrieval
Biased or Flawed? Mitigating Stereotypes in Generative Language Models by Addressing Task-Specific Flaws
Advancements and Challenges in Bangla Question Answering Models: A Comprehensive Review
2M-BELEBELE: Highly Multilingual Speech and American Sign Language Comprehension Dataset
Y-NQ: English-Yorùbá Evaluation dataset for Open-Book Reading Comprehension and Text Generation
TECO: Improving Multimodal Intent Recognition with Text Enhancement through Commonsense Knowledge Extraction
Asking Again and Again: Exploring LLM Robustness to Repeated Questions
Marco-LLM: Bridging Languages via Massive Multilingual Training for Cross-Lingual Enhancement
Mimir: Improving Video Diffusion Models for Precise Text Understanding
CPRM: A LLM-based Continual Pre-training Framework for Relevance Modeling in Commercial Search
SPAgent: Adaptive Task Decomposition and Model Selection for General Video Generation and Editing
Scholar Name Disambiguation with Search-enhanced LLM Across Language
Diffusion-Inspired Cold Start with Sufficient Prior in Computerized Adaptive Testing
Neurosymbolic Graph Enrichment for Grounded World Models
SciDQA: A Deep Reading Comprehension Dataset over Scientific Papers
Evaluating Large Language Model Capability in Vietnamese Fact-Checking Data Generation
A study of Vietnamese readability assessing through semantic and statistical features
Explaining and Improving Contrastive Decoding by Extrapolating the Probabilities of a Huge and Hypothetical LM
Diagnosing Medical Datasets with Training Dynamics
NLP and Education: using semantic similarity to evaluate filled gaps in a large-scale Cloze test in the classroom
Visualizing attention zones in machine reading comprehension models
Evaluating LLMs for Targeted Concept Simplification for Domain-Specific Texts
LLMs are Biased Evaluators But Not Biased for Retrieval Augmented Generation
Beyond Autoregression: Fast LLMs via Self-Distillation Through Time
TransformLLM: Adapting Large Language Models via LLM-Transformed Reading Comprehension Text
RoBIn: A Transformer-Based Model For Risk Of Bias Inference With Machine Reading Comprehension
Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions
Developing a Tutoring Dialog Dataset to Optimize LLMs for Educational Use
BioMistral-NLU: Towards More Generalizable Medical Language Understanding through Instruction Tuning
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.