Skip to main content
Leading NLP Ninja

Leading NLP Ninja

By jojonki

Leading NLP Ninjaでは最近のNLP (Natural Language Processing)に関連する論文をjojonkiが短く紹介します.気になったこと・質問・間違い等,フィードバック頂けると嬉しいです.
紹介する論文は,基本的に下記の論文まとめから取り上げる予定です.
github.com/jojonki/arXivNotes/issues
Available on
Apple Podcasts Logo
Castbox Logo
Google Podcasts Logo
Overcast Logo
Pocket Casts Logo
PodBean Logo
RadioPublic Logo
Spotify Logo
Currently playing episode

ep3: Large-Scale Multi-Domain Belief Tracking with Knowledge Sharing

Leading NLP NinjaAug 24, 2018

00:00
22:19
ep52(ACL): A Two-Stage Masked LM Method for Term Set Expansion
May 16, 202031:11
ep51 (arXiv): XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization
Apr 18, 202027:28
ep50 (ICLR): ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
Mar 14, 202028:02
ep49 (ICASSP): Looking Enhances Listening: Recovering Missing Speech Using Images
Feb 21, 202022:11
ep48 (AAAI): Emu: Enhancing Multilingual Sentence Embeddings with Semantic Specialization
Feb 09, 202038:34
ep47 (ICLR): ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
Jan 12, 202025:35
ep46: FreeLB: Enhanced Adversarial Training for Language Understanding
Jan 01, 202024:21
ep45: Episodic Memory in Lifelong Language Learning
Dec 01, 201936:03
ep44: 75 Languages, 1 Model: Parsing Universal Dependencies Universally
Nov 24, 201936:04
ep43: BPE-Dropout: Simple and Effective Subword Regularization
Nov 04, 201936:45
ep42: HuggingFace's Transformers: State-of-the-art Natural Language Processing
Oct 20, 201919:45
ep41: A Simple Theoretical Model of Importance for Summarization
Sep 15, 201945:26
ep40: OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs
Sep 07, 201934:17
ep39: Conversational Response Re-ranking Based on Event Causality and Role Factored Tensor Event Embedding
Aug 20, 201926:05
ep38: Trends in Natural Language Processing: ACL 2019 In Review
Aug 09, 201923:59
ep37: Multimodal Transformer Networks for End-to-End Video-Grounded Dialogue Systems
Jul 15, 201926:15
ep36: A Survey of Reinforcement Learning Informed by Natural Language
Jul 07, 201926:47
ep35: Modeling Semantic Relationship in Multi-turn Conversations with Hierarchical Latent Variables
Jun 30, 201926:37
ep34: Do Neural Dialog Systems Use the Conversation History Effectively?
Jun 07, 201916:31
ep33: Target-Guided Open-Domain Conversation
Jun 01, 201938:41
ep32: We need to talk about standard splits

ep32: We need to talk about standard splits

github上でご指摘いただきましたが,実験2のReproductionのシステムの優劣を完全に逆にして説明していることが判明しました.issueのコメント欄を御覧ください.他にも怪しくなってきた気がするのでお気付きの方はどんどんコメント頂けると嬉しいです。


第32回では,ACL 2019から,標準的なデータセットのsplitに起因するシステム比較手法の危険性を示す論文を解説しました. 今回紹介した論文はこちらのissueで解説しています. github.com/jojonki/arXivNotes/issues/241 番組への支援は,こちらからお待ちしております.

May 26, 201922:17
ep31: CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge
May 11, 201933:02
ep30: Probing the Need for Visual Context in Multimodal Machine Translation
Apr 30, 201925:32
ep29: What's in a Name? Reducing Bias in Bios without Access to Protected Attributes
Apr 22, 201931:19
ep28: Attention is not Explanation
Apr 13, 201924:54
ep27: 今年度の振り返りとこれからについて
Mar 31, 201917:15
ep26: 大規模な自動解析データが形態素解析器をどこまで小さくできるか
Mar 23, 201916:19
ep25: サブワードに基づく単語分散表現の縮約モデリング
Mar 17, 201918:23
ep24: BERT for Joint Intent Classification and Slot Filling
Mar 02, 201921:08
ep23: End-to-End Knowledge-Routed Relational Dialogue System for Automatic Diagnosis
Feb 24, 201929:19
ep22: What are the biases in my data?
Feb 17, 201913:22
ep21: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Feb 03, 201932:19
ep20: Comprehensive evaluation of statistical speech waveform synthesis
Jan 27, 201919:08
ep19: SentencePiece: A simple and language independent subword tokenizer and detokenizer for NLP
Jan 20, 201937:24
ep18: PyText: A Seamless Path from NLP research to production
Jan 14, 201928:03
ep17: User Modeling for Task Oriented Dialogues
Dec 31, 201852:40
ep16: Contextual Topic Modeling For Dialog Systems
Dec 09, 201829:08
ep15: Another Diversity-Promoting Objective Function for Neural Dialogue Generation
Dec 02, 201831:48
ep14: XNLI: Evaluating Cross-lingual Sentence Representations
Nov 25, 201837:39
ep13: You May Not Need Attention
Nov 04, 201825:36
ep12: Word Embedding based Edit Distance
Oct 28, 201830:10
ep11: Query Tracking for E-commerce Conversational Search: A Machine Comprehension Perspective
Oct 21, 201840:06
ep10: Automatic Evaluation of Neural Personality-based Chatbots
Oct 07, 201836:29
ep9: Learning and Evaluating Sparse Interpretable Sentence Embeddings
Oct 01, 201846:54
ep8: Zero-Shot Adaptive Transfer for Conversational Language Understanding
Sep 24, 201824:28
ep7: Training Millions of Personalized Dialogue Agents

ep7: Training Millions of Personalized Dialogue Agents

第7回では,EMNLP 2018でFacebookが開発したペルソナデータセット及びそのペルソナに沿ったEnd-to-end雑談対話システムを解説しました.
今回紹介した論文の解説はこちらにあります.https://github.com/jojonki/arXivNotes/issues/130
Sep 11, 201827:06
ep6: Bag of Experts Architectures for Model Reuse in Conversational Language Understanding
Sep 06, 201826:06
ep5: Analysing the potential of seq-to-seq models for incremental interpretation in task-oriented dialogue
Aug 30, 201831:47
ep4: CoQA: A Conversational Question Answering Challeng
Aug 28, 201822:14
ep3: Large-Scale Multi-Domain Belief Tracking with Knowledge Sharing
Aug 24, 201822:19