关于机器学习:2023计算机领域顶会A类以及ACL-2023自然语言处理NLP研究子方向领域汇总

31次阅读

共计 12034 个字符,预计需要花费 31 分钟才能阅读完成。

2023 年的计算语言学协会年会(ACL 2023)共蕴含 26 个畛域,代表着以后前计算语言学和自然语言解决钻研的不同方面。每个畛域都有一组相关联的关键字来形容其潜在的子畛域,这些子畛域并非排他性的,它们只形容了最受关注的子畛域,并心愿可能对该畛域蕴含的相干类型的工作提供一些更好的想法。

1. 计算机领域顶会 (A 类)

会议简称 次要畛域 会议全称 官网 截稿工夫 会议工夫
CVPR2023 计算机视觉 The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2023https://cvpr2023.thecvf.com/2022.11.112023.6.18
ICCV2023 计算机视觉 IEEE International Conference on Computer Visionhttps://iccv2023.thecvf.com/2023.3.82023.9.30
ECCV2022 计算机视觉 European Conference on Computer Visionhttps://eccv2022.ecva.net/——-2022.10.23
AAAI2023 人工智能 National Conference of the American Association for Artificial Intelligencehttps://aaai-23.aaai.org/2022.8.82023.2.7
IJCAI 2023 人工智能 National Conference of the American Association for Artificial Intelligencehttps://ijcai-22.org/#2022.8.82023.2.7
NIPS2023 机器学习 International Joint Conference on Artificial Intelligencehttps://neurips.cc/Conferences/20222023.012023.07
ICML 2023 机器学习 International Conference on Machine Learninghttps://icml.cc/2023.012023.06.24
ICLR 2023 机器学习 International Conference on Learning Representationshttps://iclr.cc/Conferences/20232022.09.212023.05.01
ICSE 2023 软件工程 International Conference on Software Engineeringhttps://conf.researchr.org/home/icse-20232022.09.012023.05.14
SIGKDD 2023 数据挖掘 ACM International Conference on Knowledge Discovery and Data Mininghttps://kdd.org/kdd2022/index.html2023.022023.08
SIGIR 2023 数据挖掘 ACM International Conference on Research and Development in Information Retrievalhttps://sigir.org/sigir2022/2023.012023.07
ACL 2023 计算语言 Association of Computational Linguisticshttps://www.2022.aclweb.org/2022.112023.05
ACM MM 2023 多媒体 ACM International Conference on Multimediahttps://2023.acmmmsys.org/participation/important-dates/2022.11.182023.6.7
WWW2023 网络应用 International World Wide Web Conferencehttps://www2023.thewebconf.org/2022.10.62023.05.01
SIGGRAPH 2023 图形学 ACM SIG International Conference on Computer Graphics and Interactive Techniqueshttps://s2022.siggraph.org/2023.012023.08
CHI 2023 人机交互 ACM Conference on Human Factors in Computing Systemshttps://chi2023.acm.org/2022.09.082023.04.23
CSCW 2023 人机交互 ACM Conference on Computer Supported Cooperative Work and Social Computinghttps://cscw.acm.org/2023/2023.01.152023.10.13
CCS 2023 信息安全 ACM Conference on Computer and Communications Securityhttps://www.sigsac.org/ccs/CCS2022/2023.012023.11
VLDB 2023 数据管理 International Conference on Very Large Data Baseshttps://www.vldb.org/2023/?submission-guidelines2023.03.012023.08.28
STOC 2023 计算机实践 ACM Symposium on the Theory of Computinghttp://acm-stoc.org/stoc2022/2022.112023.06

2.ACL 2023 自然语言解决(NLP)钻研子方向畛域汇总

(一)计算社会科学和文化剖析 (Computational Social Science and Cultural Analytics)

  1. 人类行为剖析 (Human behavior analysis)
  2. 态度检测 (Stance detection)
  3. 框架检测和剖析 (Frame detection and analysis)
  4. 怨恨舆论检测 (Hate speech detection)
  5. 错误信息检测和剖析 (Misinformation detection and analysis)
  6. 人口心理画像预测 (psycho-demographic trait prediction)
  7. 情绪检测和剖析 (emotion detection and analysis)
  8. 表情符号预测和剖析 (emoji prediction and analysis)
  9. 语言和文化偏见剖析 (language/cultural bias analysis)
  10. 人机交互 (human-computer interaction)
  11. 社会语言学 (sociolinguistics)
  12. 用于社会剖析的自然语言解决工具 (NLP tools for social analysis)
  13. 新闻和社交媒体的定量分析 (quantiative analyses of news and/or social media)

(二)对话和交互零碎 (Dialogue and Interactive Systems)

  1. 书面语对话零碎 (Spoken dialogue systems)
  2. 评估指标 (Evaluation and metrics)
  3. 工作型 (Task-oriented)
  4. 人工染指 (Human-in-a-loop)
  5. 偏见和毒性 (Bias/toxity)
  6. 事实性 (Factuality)
  7. 检索 (Retrieval)
  8. 常识加强 (Knowledge augmented)
  9. 常识推理 (Commonsense reasoning)
  10. 互动讲故事 (Interactive storytelling)
  11. 具象代理人 (Embodied agents)
  12. 利用 (Applications)
  13. 多模态对话零碎 (Multi-modal dialogue systems)
  14. 常识驱动对话 (Grounded dialog)
  15. 多语言和低资源 (Multilingual / low-resource)
  16. 对话状态追踪 (Dialogue state tracking)
  17. 对话建模 (Conversational modeling)

(三)话语和语用学 (Discourse and Pragmatics)

  1. 回指消解 (Anaphora resolution)
  2. 共指消解 (Coreference resolution)
  3. 桥接消解 (Bridging resolution)
  4. 连贯 (Coherence)
  5. 统一 (Cohesion)
  6. 话语关系 (Discourse relations)
  7. 话语分析 (Discourse parsing)
  8. 对话 (Dialogue)
  9. 会话 (Conversation)
  10. 话语和多语性 (Dialugue and multilinguality)
  11. 观点开掘 (Argument mining)
  12. 交际 (Communication)

(四)自然语言解决和伦理 (Ethics and NLP)

  1. 数据伦理 (Data ethics)
  2. 模型偏见和公正性评估 (Model bias/fairness evaluation)
  3. 缩小模型的偏见和不公平性 (Model bias/unfairness mitigation)
  4. 自然语言解决中的人类因素 (Human factors in NLP)
  5. 参加式和基于社群的自然语言解决 (Participatory/community-based NLP)
  6. 自然语言解决利用中的道德思考 (Ethical considerations in NLP)
  7. 透明性 (Transparency)
  8. 政策和治理 (Policy and governance)
  9. 观点和批评 (Reflections and critiques)

(五)语言生成 (Generation)

  1. 人工评估 (Human evaluation)
  2. 主动评估 (Automatic evaluation)
  3. 多语言 (Multilingualism)
  4. 高效模型 (Efficient models)
  5. 少样本生成 (Few-shot generation)
  6. 剖析 (Analysis)
  7. 畛域适应 (Domain adaptation)
  8. 数据到文本生成 (Data-to-text generation)
  9. 文本到文博生成 (Text-to-text generation)
  10. 推断办法 (Inference methods)
  11. 模型构造 (Model architectures)
  12. 检索加强生成 (Retrieval-augmented generation)
  13. 交互和单干生成 (Interactive and collaborative generation)

(六)信息抽取 (Information Extraction)

  1. 命名实体辨认和关系抽取 (Named entity recognition and relation extraction)
  2. 事件抽取 (Event extraction)
  3. 凋谢信息抽取 (Open information extraction)
  4. 知识库构建 (Knowledge base construction)
  5. 实体连贯和消歧 (Entity linking and disambiguation)
  6. 文档级抽取 (Document-level extraction)
  7. 多语言抽取 (Multilingual extraction)
  8. 小样本和零样本抽取 (Zero-/few-shot extraction)

(七)信息检索和文本开掘 (Information Retrieval and Text Mining)

  1. 段落检索 (Passage retrieval)
  2. 密集检索 (Dense retrieval)
  3. 文档表征 (Document representation)
  4. 哈希 (Hashing)
  5. 重排序 (Re-ranking)
  6. 预训练 (Pre-training)
  7. 比照学习 (Constrastive learning)

(八)自然语言解决模型的可解释性与剖析 (Interpretability and Analysis of Models in NLP)

  1. 对抗性攻打 / 例子 / 训练 (Adversarial attacks/examples/training)
  2. 校对和不确定性 (Calibration/uncertainty)
  3. 反事实和比照解释 (Counterfactual/contrastive explanations)
  4. 数据影响 (Data influence)
  5. 数据瑕疵 (Data shortcuts/artifacts)
  6. 解释的忠诚度 (Explantion faithfulness)
  7. 特色归因 (Feature attribution)
  8. 自在文本和自然语言解释 (Free-text/natural language explanation)
  9. 样本硬度 (Hardness of samples)
  10. 构造和概念解释 (Hierarchical & concept explanations)
  11. 以人为主体的利用评估 (Human-subject application-grounded evaluations)
  12. 常识追溯、发现和推导 (Knowledge tracing/discovering/inducing)
  13. 探索 (Probing)
  14. 稳健性 (Robustness)
  15. 话题建模 (Topic modeling)

(九)视觉、机器人等畛域的语言根底 (Language Grounding to Vision, Robotics and Beyond)

  1. 视觉语言导航 (Visual Language Navigation)
  2. 跨模态预训练 (Cross-modal pretraining)
  3. 图像文本匹配 (Image text macthing)
  4. 跨模态内容生成 (Cross-modal content generation)
  5. 视觉问答 (Visual question answering)
  6. 跨模态利用 (Cross-modal application)
  7. 跨模态信息抽取 (Cross-modal information extraction)
  8. 跨模态机器翻译 (Cross-modal machine translation)

(十)大模型 (Large Language Models)

  1. 预训练 (Pre-training)
  2. 提醒 (Prompting)
  3. 规模化 (Scaling)
  4. 稠密模型 (Sparse models)
  5. 检索加强模型 (Retrieval-augmented models)
  6. 伦理 (Ethics)
  7. 可解释性和剖析 (Interpretability/Analysis)
  8. 间断学习 (Continual learning)
  9. 平安和隐衷 (Security and privacy)
  10. 利用 (Applications)
  11. 稳健性 (Robustness)
  12. 微调 (Fine-tuning)

(十一)语言多样性 (Language Diversity)

  1. 少资源语言 (Less-resource languages)
  2. 濒危语言 (Endangered languages)
  3. 土著语言 (Indigenous languages)
  4. 少数民族语言 (Minoritized languages)
  5. 语言记录 (Language documentation)
  6. 少资源语言的资源 (Resources for less-resourced languages)
  7. 软件和工具 (Software and tools)

(十二)语言学实践、认知建模和心理语言学 (Linguistic Theories, Cognitive Modeling and Psycholinguistics)

  1. 语言学实践 (Linguistic theories)
  2. 认知建模 (Cognitive modeling)
  3. 计算心理语言学 (Computational pyscholinguistics)

(十三)自然语言解决中的机器学习 (Machine Learning for NLP)

  1. 基于图的办法 (Graph-based methods)
  2. 常识加强的办法 (Knowledge-augmented methods)
  3. 多任务学习 (Multi-task learning)
  4. 自监督学习 (Self-supervised learning)
  5. 比照学习 (Contrastive learning)
  6. 生成模型 (Generation model)
  7. 数据加强 (Data augmentation)
  8. 词嵌入 (Word embedding)
  9. 结构化预测 (Structured prediction)
  10. 迁徙学习和畛域适应 (Transfer learning / domain adaptation)
  11. 表征学习 (Representation learning)
  12. 泛化 (Generalization)
  13. 模型压缩办法 (Model compression methods)
  14. 参数高效的微调办法 (Parameter-efficient finetuning)
  15. 少样本学习 (Few-shot learning)
  16. 强化学习 (Reinforcement learning)
  17. 优化办法 (Optimization methods)
  18. 间断学习 (Continual learning)
  19. 反抗学习 (Adversarial training)
  20. 元学习 (Meta learning)
  21. 因果关系 (Causality)
  22. 图模型 (Graphical models)
  23. 人参加的学习和被动学习 (Human-in-a-loop / Active learning)

(十四)机器翻译 (Machine Translation)

  1. 主动评估 (Automatic evaluation)
  2. 偏见 (Biases)
  3. 畛域适应 (Domain adaptation)
  4. 机器翻译的高效推理方法 (Efficient inference for MT)
  5. 高效机器翻译训练 (Efficient MT training)
  6. 少样本和零样本机器翻译 (Few-/Zero-shot MT)
  7. 人工评估 (Human evaluation)
  8. 交互机器翻译 (Interactive MT)
  9. 机器翻译部署和保护 (MT deployment and maintainence)
  10. 机器翻译实践 (MT theory)
  11. 建模 (Modeling)
  12. 多语言机器翻译 (Multilingual MT)
  13. 多模态 (Multimodality)
  14. 机器翻译的线上使用 (Online adaptation for MT)
  15. 并行解码和非自回归的机器翻译 (Parallel decoding/non-autoregressive MT)
  16. 机器翻译预训练 (Pre-training for MT)
  17. 规模化 (Scaling)
  18. 语音翻译 (Speech translation)
  19. 转码翻译 (Code-switching translation)
  20. 词表学习 (Vocabulary learning)

(十五)多语言和跨语言自然语言解决 (Multilingualism and Cross-Lingual NLP)

  1. 转码 (Code-switching)
  2. 混合语言 (Mixed language)
  3. 多语言 (Multilingualism)
  4. 语言接触 (Language contact)
  5. 语言变迁 (Language change)
  6. 语言变体 (Language variation)
  7. 跨语言迁徙 (Cross-lingual transfer)
  8. 多语言表征 (Multilingual representation)
  9. 多语言预训练 (Multilingual pre-training)
  10. 多语言基线 (Multilingual benchmark)
  11. 多语言评估 (Multilingual evaluation)
  12. 方言和语言变种 (Dialects and language varieties)

(十六)自然语言解决利用 (NLP Applications)

  1. 教育利用、语法纠错、文章打分 (Educational applications, GEC, essay scoring)
  2. 怨恨舆论检测 (Hate speech detection)
  3. 多模态利用 (Multimodal applications)
  4. 代码生成和了解 (Code generation and understanding)
  5. 事实检测、流言和错误信息检测 (Fact checking, rumour/misinformation detection)
  6. 医疗利用、诊断自然语言解决 (Healthcare applications, clinical NLP)
  7. 金融和商务自然语言解决 (Financial/business NLP)
  8. 法律自然语言解决 (Legal NLP)
  9. 数学自然语言解决 (Mathematical NLP)
  10. 平安和隐衷 (Security/privacy)
  11. 历史自然语言解决 (Historical NLP)
  12. 常识图谱 (Knowledge graph)

(十七)音系学、形态学和词语宰割 (Phonology, Morphology and Word Segmentation)

  1. 状态变动 (Morphological inflection)
  2. 范式演绎 (Paradigm induction)
  3. 形态学宰割 (Morphological segementation)
  4. 子词表征 (Subword representations)
  5. 中文宰割 (Chinese segmentation)
  6. 词性还原 (Lemmatization)
  7. 有限元形态学 (Finite-state morphology)
  8. 形态学剖析 (Morphological analysis)
  9. 音系学 (Phonology)
  10. 字素音素转换 (Grapheme-to-phoneme conversion)
  11. 发音建模 (Pronunciation modeling)

(十八)问答 (Question Answering)

  1. 常识问答 (Commonsense QA)
  2. 浏览了解 (Reading comprehension)
  3. 逻辑推理 (Logic reasoning)
  4. 多模态问答 (Multimodal QA)
  5. 知识库问答 (Knowledge base QA)
  6. 语义剖析 (Semantic parsing)
  7. 多跳问答 (Multihop QA)
  8. 生物医学问答 (Biomedical QA)
  9. 多语言问答 (Multilingual QA)
  10. 可解释性 (Interpretability)
  11. 泛化 (Generalization)
  12. 推理 (Reasoning)
  13. 对话问答 (Conversational QA)
  14. 少样本问答 (Few-shot QA)
  15. 数学问答 (Math QA)
  16. 表格问答 (Table QA)
  17. 凋谢域问答 (Open-domain QA)
  18. 问题生成 (Question generation)

(十九)语言资源及评估 (Resources and Evaluation)

  1. 语料库构建 (Corpus creation)
  2. 基线构建 (Benchmarking)
  3. 语言资源 (Language resources)
  4. 多语言语料库 (Multilingual corpora)
  5. 词表构建 (Lexicon creation)
  6. 语言资源的主动构建与评估 (Automatic creation and evaluation of language
    resources)
  7. 自然语言解决数据集 (NLP datasets)
  8. 数据集主动评估 (Automatic evaluation of datasets)
  9. 评估办法 (Evaluation methodologies)
  10. 低资源语言数据集 (Datasets for low resource languages)
  11. 测量指标 (Metrics)
  12. 复现性 (Reproducibility)
  13. 用于评估的统计测验 (Statistical testing for evaluation)

(二十)语义学:词汇层面 (Semantics: Lexical)

  1. 一词多义 (Polysemy)
  2. 词汇关系 (Lexical relationships)
  3. 文本蕴含 (Textual entailment)
  4. 语义合成性 (Compositionality)
  5. 多词表白 (Multi-word expressions)
  6. 同义转换 (Paraphrasing)
  7. 隐喻 (Metaphor)
  8. 词汇语义变迁 (Lexical semantic change)
  9. 词嵌入 (Word embeddings)
  10. 认知 (Cognition)
  11. 词汇资源 (Lexical resources)
  12. 情感剖析 (Sentiment analysis)
  13. 多语性 (Multilinguality)
  14. 可解释性 (Interpretability)
  15. 探索性钻研 (Probing)

(二十一)语义学:句级语义、文本推断和其余畛域 (Semantics: Sentence-Level Semantics, Textual Inference and Other Areas)

  1. 同义句辨认 (Paraphrase recognition)
  2. 文本蕴含 (Textual entailment)
  3. 自然语言推理 (Natural language inference)
  4. 逻辑推理 (Reasoning)
  5. 文本语义相似性 (Semantic textual similarity)
  6. 短语和句子嵌入 (Phrase/sentence embedding)
  7. 同义句生成 (Paraphrase generation)
  8. 文本简化 (Text simiplification)
  9. 词和短语对齐 (Word/phrase alignment)

(二十二)情感剖析、文本格调剖析和论点开掘 (Sentiment Analysis, Stylistic Analysis and Argument Mining)

  1. 论点开掘 (Argument mining)
  2. 观点检测 (Stance detection)
  3. 论点品质评估 (Argument quality assessment)
  4. 修辞和框架 (Rhetoric and framing)
  5. 论证计划和推理 (Argument schemes and reasoning)
  6. 论点生成 (Argument generation)
  7. 格调剖析 (Style analysis)
  8. 格调生成 (Style generation)
  9. 利用 (Applications)

(二十三)语音和多模态 (Speech and Multimodality)

  1. 主动语音辨认 (Automatic speech recognition)
  2. 书面语语言了解 (Spoken language understanding)
  3. 口语翻译 (Spoken language translation)
  4. 书面语语言根底 (Spoken language grounding)
  5. 语音和视觉 (Speech and vision)
  6. 书面语查问问答 (QA via spoken queries)
  7. 书面语对话 (Spoken dialog)
  8. 视频解决 (Video processing)
  9. 语音根底 (Speech technologies)
  10. 多模态 (Multimodality)

(二十四)文摘 (Summarization)

  1. 抽取文摘 (Extractive summarization)
  2. 摘要文摘 (Abstractive summarization)
  3. 多模态文摘 (Multimodal summarization)
  4. 多语言文摘 (Multilingual summarization)
  5. 对话文摘 (Conversational summarization)
  6. 面向查问的文摘 (Query-focused summarization)
  7. 多文档文摘 (Multi-document summarization)
  8. 长格局文摘 (Long-form summarization)
  9. 句子压缩 (Sentence compression)
  10. 少样本文摘 (Few-shot summarization)
  11. 构造 (Architectures)
  12. 评估 (Evaluation)
  13. 事实性 (Factuality)

(二十五)句法学:标注、组块剖析和句法分析 (Syntax: Tagging, Chunking and Parsing)

  1. 组块剖析、浅层剖析 (Chunking, shallow-parsing)
  2. 词性标注 (Part-of-speech tagging)
  3. 依存句法分析 (Dependency parsing)
  4. 成分句法分析 (Constituency parsing)
  5. 深层句法分析 (Deep syntax parsing)
  6. 语义剖析 (Semantic parsing)
  7. 句法语义接口 (Syntax-semantic inferface)
  8. 状态句法相干工作的标注和数据集 (Optimized annotations or data set for morpho-syntax
    related tasks) 句法分析算法 (Parsing algorithms)
  9. 语法和基于常识的办法 (Grammar and knowledge-based approach)
  10. 多任务办法 (Multi-task approaches)
  11. 面向大型多语言的办法 (Massively multilingual oriented approaches)
  12. 低资源语言词性标注、句法分析和相干工作 (Low-resource languages pos-tagging, parsing
    and related tasks)
  13. 状态丰盛语言的词性标注、句法分析和相干工作 (Morphologically-rich languages pos tagging,
    parsing and related tasks)

(二十六)主题畛域:事实检测 (Theme Track: Reality Check)

  1. 因为谬误的起因而正确 (Right for the wrong reasons)
  2. 理论使用中的教训 (Lessons from deployment)
  3. (非)泛化能力 [(Non-)generalization]
  4. (非)复现能力 [(Non-)reproducibility)]
  5. 评估 (Evaluation)
  6. 办法 (Methodology)
  7. 负面后果 (Negative results)
  8. 人工智能噱头和期待 (AI hype and expectations)
  9. 迷信 vs 工程 (Science-vs-engineering)
  10. 其余畛域的联合 (Lessons from other fields)
正文完
 0