當(dāng)前位置：首頁(yè) > 编程资源 > 编程问答 >内容正文

编程问答

视频问答兴起，多跳问答热度衰退，92篇论文看智能问答的发展趋势

發(fā)布時(shí)間：2024/7/5 编程问答 51 豆豆

生活随笔收集整理的這篇文章主要介紹了视频问答兴起，多跳问答热度衰退，92篇论文看智能问答的发展趋势小編覺(jué)得挺不錯(cuò)的,現(xiàn)在分享給大家,幫大家做個(gè)參考.

星標(biāo)/置頂小屋，帶你解鎖

最萌最前沿的NLP、搜索與推薦技術(shù)

文 | 舒意恒（南京大學(xué)碩士生，知識(shí)圖譜方向）

編 |? 北大小才女小軼

2019年的時(shí)候，舒意恒Y.Shu整理了一份《2019年，智能問(wèn)答有哪些研究方向？》，如今2020年已經(jīng)過(guò)去了一半，該領(lǐng)域的研究趨勢(shì)發(fā)生了一些有趣的變化。于是Y.Shu繼續(xù)為大家整理了今年ACL、AAAI、IJCAI、SIGIR、CVPR、ICML、KDD等頂級(jí)會(huì)議上智能問(wèn)答方向的92篇論文，進(jìn)行了分方向梳理，并為大家打包了這92篇的論文PDF掛在訂閱號(hào)后臺(tái)方便大家下載。

在智能問(wèn)答涉及的各個(gè)小方向上，視覺(jué)/視頻問(wèn)答、對(duì)話、問(wèn)題生成等主題成為最熱門的研究方向。不同于關(guān)于 2019 年 QA 研究的統(tǒng)計(jì)，多跳問(wèn)答在今年上半年似乎沒(méi)有持續(xù)的熱度。

在訂閱號(hào) 「夕小瑤的賣萌屋」 后臺(tái)回復(fù)關(guān)鍵詞【0713】，即可下載92篇論文pdf合集。

基準(zhǔn)測(cè)試

AAAI 2020｜Getting Closer to AI Complete Question Answering: A Set of Prerequisite Real Tasks 新的閱讀理解數(shù)據(jù)集，包含基于文本的世界知識(shí)和無(wú)法回答的問(wèn)題
ACL 2020｜What Question Answering can Learn from Trivia Nerds 問(wèn)答系統(tǒng)能從冷知識(shí)書(shū)呆子那兒學(xué)到啥？

常識(shí)問(wèn)答

AAAI 2020｜Graph-Based Reasoning over Heterogeneous External Knowledge for Commonsense Question Answering
AAAI 2020｜PIQA: Reasoning about Physical Commonsense in Natural Language
ProtoQA: A Question Answering Dataset for Prototypical Common-Sense Reasoning 常識(shí)推理數(shù)據(jù)集

社區(qū)問(wèn)答

AAAI 2020｜Attentive User-Engaged Adversarial Neural Network for Community Question Answering
AAAI 2020｜Joint Learning of Answer Selection and Answer Summary Generation in Community Question Answering

文檔級(jí)問(wèn)答

ACL 2020｜Probabilistic Assumptions Matter: Improved Models for Distantly-Supervised Document-Level Question Answering
ACL 2020｜RikiNet: Reading Wikipedia Pages for Natural Question Answering 長(zhǎng)文檔閱讀

片段選擇

ACL 2020｜Transformers to Learn Hierarchical Contexts in Multiparty Dialogue for Span-based Question Answering
ACL 2020｜Span Selection Pre-training for Question Answering
ACL 2020｜CorefQA: Coreference Resolution as Query-based Span Prediction 基于片段預(yù)測(cè)的共指消解

開(kāi)放域問(wèn)答

ACL 2020｜Contextualized Sparse Representations for Real-Time Open-Domain Question Answering
ACL 2020｜Dense Passage Retrieval for Open-Domain Question Answering
ACL 2020｜Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index

知識(shí)庫(kù)問(wèn)答與多跳問(wèn)答

ACL 2020｜Improving Multi-hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings 利用知識(shí)圖譜嵌入的多跳問(wèn)答模型，獲得 SOTA
ACL 2020｜Unsupervised Alignment-based Iterative Evidence Retrieval for Multi-hop Question Answering 多跳問(wèn)答的迭代解釋檢索
ACL 2020｜Query Graph Generation for Answering Multi-hop Complex Questions from Knowledge Bases 知識(shí)庫(kù)問(wèn)答的多跳復(fù)雜問(wèn)題查詢圖生成
AAAI 2020｜Skeleton-based Semantic Parsing for Complex Questions over Knowledge Bases

高效的問(wèn)答系統(tǒng)

ACL 2020｜DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering 問(wèn)題和文段分別編碼
ACL 2020｜Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index

對(duì)話

AAAI 2020｜Improving Knowledge-Aware Dialogue Generation via Knowledge Base Question Answering
AAAI 2020｜An Empirical Study of Content Understanding in Conversational Question Answering
SIGIR 2020｜Open-Retrieval Conversational Question Answering
ACL 2020｜Fluent Response Generation for Conversational Question Answering
ACL 2020｜Do not let the history haunt you: Mitigating Compounding Errors in Conversational Question Answering
ACL 2020｜Learning to Identify Follow-Up Questions in Conversational Question Answering
ACL 2020｜DoQA - Accessing Domain-Specific FAQs via Conversational QA

視覺(jué)問(wèn)答

包括針對(duì)圖像、視頻的問(wèn)答。

IJCAI 2020｜Mucko: Multi-Layer Cross-Modal Knowledge Reasoning for Fact-based Visual Question Answering
ICML 2020｜Probing Emergent Semantics in Predictive Agents via Question Answering
IEEE TGRS｜RSVQA: Visual Question Answering for Remote Sensing Data
CVPR 2020｜Counterfactual Samples Synthesizing for Robust Visual Question Answering
AAAI 2020 workshop｜A Study on Multimodal and Interactive Explanations for Visual Question Answering
CVPR 2020 oral｜Hierarchical Conditional Relation Networks for Video Question Answering
CVPR 2020｜On the General Value of Evidence, and Bilingual Scene-Text Visual Question Answering
IEEE SEMANTIC COMPUTING｜Augmenting Visual Question Answering with Semantic Frame Information in a Multitask Learning Approach
AAAI 2020｜Re-Attention for Visual Question Answering
AAAI 2020｜Divide and Conquer: Question-Guided Spatio-Temporal Contextual Attention for Video Question Answering
AAAI 2020｜Segment-then-Rank: Non-factoid Question Answering on Instructional Videos
AAAI 2020｜Location-aware Graph Convolutional Networks for Video Question Answering
AAAI 2020｜Multi-Question Learning for Visual Question Answering
AAAI 2020｜Reasoning with Heterogeneous Graph Alignment for Video Question Answering
AAAI 2020｜Unified Vision-Language Pre-Training for Image Captioning and VQA
AAAI 2020｜KnowIT VQA: Answering Knowledge-Based Questions about Videos
AAAI 2020｜Overcoming Language Priors in VQA via Decomposed Linguistic Representations
AAAI 2020｜Explanation vs Attention: A Two-Player Game to Obtain Attention for VQA
ACL 2020｜Aligned Dual Channel Graph Convolutional Network for Visual Question Answering
ACL 2020｜Multimodal Neural Graph Memory Networks for Visual Question Answering
ACL 2020｜TVQA+: Spatio-Temporal Grounding for Video Question Answering
ACL 2020｜Dense-Caption Matching and Frame-Selection Gating for Temporal Localization in VideoQA
ACL 2020｜A negative case analysis of visual grounding methods for VQA

問(wèn)題生成

ACL 2020 workshop｜A Question Type Driven and Copy Loss Enhanced Frameworkfor Answer-Agnostic Neural Question Generation
ACL 2020｜Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs
ACL 2020｜On the Importance of Diversity in Question Generation for QA
AAAI 2020｜Improving Question Generation with Sentence-level Semantic Matching and Answer Position Inferring
AAAI 2020｜Conclusion-Supplement Answer Generation for Non-Factoid Questions
AAAI 2020｜Visual Dialogue State Tracking for Question Generation
AAAI 2020｜Neural Question Generation with Answer Pivot
AAAI 2020｜Capturing Greater Context for Question Generation

數(shù)據(jù)集

值得注意的是，以下一些文章未發(fā)現(xiàn)正式發(fā)表在會(huì)議/期刊上。

RuBQ: A Russian Dataset for Question Answering over Wikidata 俄語(yǔ)問(wèn)答數(shù)據(jù)集
ProtoQA: A Question Answering Dataset for Prototypical Common-Sense Reasoning 常識(shí)推理數(shù)據(jù)集
Event-QA: A Dataset for Event-Centric Question Answering over Knowledge Graphs
HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data
FQuAD: French Question Answering Dataset
AAAI 2020｜JEC-QA: A Legal-Domain Question Answering Dataset
AAAI 2020｜QASC: A Dataset for Question Answering via Sentence Composition
ACL 2020｜The TechQA Dataset 包含技術(shù)論壇上的實(shí)際問(wèn)題的數(shù)據(jù)集
ACL 2020｜Controlled Crowdsourcing for High-Quality QA-SRL Annotation 其中 SRL 指語(yǔ)義角色標(biāo)簽
AAAI 2020｜How to Ask Better Questions? A Large-Scale Multi-Domain Dataset for Rewriting Ill-Formed Questions

表格問(wèn)答

AAAI 2020｜CFGNN: Cross Flow Graph Neural Networks for Question Answering on Complex Tables
HybridQA: A Dataset of Multi-Hop Question Answering over Tabular and Textual Data

無(wú)監(jiān)督 QA

ACL 2020｜Harvesting and Refining Question-Answer Pairs for Unsupervised QA
ACL 2020｜Template-Based Question Generation from Retrieved Sentences for Improved Unsupervised Question Answering
Facebook｜Unsupervised Question Decomposition for Question Answering

垂直領(lǐng)域問(wèn)答

ACL 2020 workshop｜Entity-Enriched Neural Models for Clinical Question Answering
ACL 2020｜Talk to Papers: Bringing Neural Question Answering to Academic Search
IEEE TMI｜A Question-Centric Model for Visual Question Answering in Medical Imaging
ACL 2020｜Clinical Reading Comprehension: A Thorough Analysis of the emrQA Dataset
AAAI 2020｜On the Generation of Medical Question-Answer Pairs
AAAI 2020｜Iteratively Questioning and Answering for Interpretable Legal Judgment Prediction

對(duì)問(wèn)答系統(tǒng)的評(píng)估/利用問(wèn)答評(píng)估

ACL 2020｜MultiReQA: A Cross-Domain Evaluation for Retrieval Question Answering Models
ACL 2020｜FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization
ACL 2020｜Asking and Answering Questions to Evaluate the Factual Consistency of Summaries
ACL 2020｜MLQA: Evaluating Cross-lingual Extractive Question Answering

問(wèn)答系統(tǒng)的解釋

ACL 2020｜Unsupervised Alignment-based Iterative Evidence Retrieval for Multi-hop Question Answering 多跳問(wèn)答的迭代解釋檢索

其他

ACL 2020｜Selective Question Answering under Domain Shift
KDD 2020｜Mining Implicit Relevance Feedback from User Behavior for Web Question Answering
ACL 2020｜Logic-Guided Data Augmentation and Regularization for Consistent Question Answering
AAAI 2020｜Knowledge and Cross-Pair Pattern Guided Semantic Matching for Question Answering
AAAI 2020｜ManyModalQA: Modality Disambiguation and QA over Diverse Inputs 多模態(tài) QA
ACL 2020｜A Methodology for Creating Question Answering Corpora Using Inverse Data Annotation
ACL 2020｜Multi-source Meta Transfer for Low Resource Multiple-Choice Question Answering

好啦，只能幫大家到這里了。