自然语言处理

大语言模型LLM

kaggle-LLM Science Exam

★★★★★

• wikipedia相关知识内容, 由给定prompt问题和五个选项，预测正确答案

• 关键在于RAG优化, 主流包括采用LLM做或者deberta分类来做

  • 标题和第一句话来召回文章articles，再从文章中召回相关句子

  • 或直接召回整个段落, (separated by '\n')

• 1st inference code discussion

• 3rd rerank code discussion

• 6th code discussion

• 12nd discussion-zhihu

• example: Elastic search in kaggle

• kaggle LLM science大模型比赛金牌方案总结 - 包包大人的文章 - 知乎

• kaggle大模型竞赛优胜方案总结与思考 - hxshine的文章 - 知乎

kaggle-LLM-Detect AI Generated Text

★★★★☆

• 1st code

LMSYS - Chatbot Arena Human Preference Predictions

• 1st code

• 2nd code

• 3rd discussion

WSDM 2024 Conversational Multi-Doc QA

• 1st code

KDD-Meta_RAG

• code

• code

• code

KDD-Amazon

• code

kaggle-AI Mathematical Olympiad

• 1st infer code

• 2nd code discussion

• 3rd infer code discussion

天池-SMP 2023 ChatGLM金融大模型挑战赛

★★★★★

• 包含pdf解析、SQL生成、意图识别、RAG召回、prompt问答

• all top code

2023全球智能汽车AI挑战赛

★★★☆☆

• PDF解析, RAG, LLM prompt问答

• 4th code

• 6th discussion

• 13rd code

ATEC2023-科技精英赛—大模型的知识引入

• 7th code

大语言模型微调数据竞赛

• 2nd code

CCF AIOps 2024

• 1st discussion

• 2nd code

text classification 文本分类/回归

Feedback Prize - English Language Learning

• 1st code discussion

• 3rd code discussion

• 13rd code

kaggle-CommonLit Readability Prize

• 1st code discussion

• 2nd code discussion

• 3rd code discussion

• 4th code discussion

• 5th discussion

• 6th discussion

• 9th discussion

• 12th discussion

• 13th discussion

• 14th discussion

• 15th discussion

• 16th discussion

• 18th discussion

• 20th code discussion

• 22nd discussion

• 24th discussion

• 25th. discussion

• 28th. discussion

• 37th. discussion

• 40th. discussion

• 47th. discussion

• 48th. discussion

• 69th. discussion

• 87th. discussion

• 100th. discussion

• code

kaggle2022-Jigsaw Rate Severity of Toxic Comments

• 1st code

• 14th code discussion

kaggle2020-Jigsaw Multilingual Toxic Comment Classification

• 1st discussion post-processing

• 3rd code discussion

• 4th discussion

• 6th discussion

• 10th discussion

kaggle2019-Jigsaw Unintended Bias in Toxicity Classification

• 3rd code

• 4th code

• 7th code

• 10th code

• 18th code

kaggle2018-Toxic Comment Classification Challenge

• 19th code

• 34th code

CCF-BDCI 2019 互联网新闻情感分析

• code

kaggle-Google QUEST Q&A Labeling

• 1st code discussion

---

named entity recognition 实体命名识别

kaggle-NBME

• baseline train inference

• baseline train inference

kaggle-feedback

★★★★★

• 1st code discussion

• 2nd code discussion

• 3rd code discussion

• 4th code discussion

• 5th code discussion

• 9th code discussion

• 11st code discussion

• baseline

  • https://github.com/abhishekkrthakur/long-text-token-classification

  • https://www.kaggle.com/code/cdeotte/pytorch-bigbird-ner-cv-0-615

  • https://www.kaggle.com/code/cdeotte/tensorflow-longformer-ner-cv-0-633

人工智能技术创新大赛——商品标题实体识别

• 小结

CCKS2019-中文短文本的实体链指

• 1st code

• unk code

天池中药说明书实体识别挑战

• 1st code

• 4th code

达观杯

• code

• code

• code

CCF金融信息负面及主体判定

• https://github.com/xiong666/ccf_financial_negative

• https://github.com/Chevalier1024/CCF-BDCI-ABSA

• https://github.com/rebornZH/2019-CCF-BDCI-NLP

---

text match 文本匹配

标签传播/Siamese RNN

kaggle-U.S. Patent Phrase to Phrase Matching

• 1st code discussion

• 2nd discussion

• 3rd discussion

• 5th discussion

• 7th discussion

• 8th discussion

• 10th discussion

• 12th discussion

• 24th discussion

• 31st code discussion

• 41st discussion

• Kaggle 专利匹配比赛金牌方案赛后总结 - 致Great的文章 - 知乎

kaggle-TensorFlow 2.0 Question Answering

• 7st code

• collection dicsussion

CCF-房产行业聊天问答匹配

• 1st code discussion

• 24th code

• unk code

天池-新冠疫情相似句对判定大赛

• 1st code

• 4th code

• 6th code discussion

• 8th code

Kaggle: Quora Question Pairs

• 1st code discussion

• 4th code

• 128th code

• http://www.wuyuanhao.com/2019/02/25/quora-insincere-questions-classification%e6%80%bb%e7%bb%93/

CCF-技术需求与技术成果项目之间关联度

• 1st code

天池-小布助手对话短文本语义匹配

• 1st code

• 4th code

sohu-文本匹配

• 2nd code

• https://zhuanlan.zhihu.com/p/533808475

新冠疫情相似句对判定大赛

• https://github.com/zzy99/epidemic-sentence-pair

• https://github.com/daniellibin/nCoV-2019-sentence-similarity

• https://mp.weixin.qq.com/s/B267GHm16ZIlKkxhOJmMqg

• https://github.com/Rowchen/pytorch-for-Text-Matching

---

信息检索

kaggle-Learning Equality - Curriculum Recommendations

★★★★★

• 多语言content与topic的文本匹配问题

• 召回：tfidf + transformer arcface + rule

• 排序：LGB

• 1st code discussion

• 2nd code discussion

• 3rd code discussion

• 4th code discussion

• 5th discussion

• 6th discussion

• 7th discussion

• 9th discussion

• 10th discussion

• 12nd discussion

• 15th discussion

• 26th discussion

EEDI

• 1st [code](https://github.com/rbiswasfc/eedi-mining-misconceptions

• 10th [code](https://github.com/shyoulala/Kaggle_Eedi_2024_sayoulala

天池-问天引擎电商搜索算法赛

★★★☆☆

• 电商场景下，中文匹配. 开源较少

• 2nd code discussion

• 13rd code

wsdm2020

★★★★☆

• 1st code

• 2nd code

• 3rd code

• 4th code

• 8th code

• 9th code

kaggle-AI4code

★★★★★

• notebook根据code cell 顺序markdown cell顺序

• 1st code

• code

• https://github.com/louis-she/ai4code

• https://www.kaggle.com/code/nickuzmenkov/ai4code-tensorflow-distilbert-baseline

WSDM2023 Pre-training for Web Search

• 1st code

**KDD Cup 2024-AQA

• 1st code

  • 多fold来分别做这件事。多步迭代选择困难负样本，最终选择Top20. 初始向量模型召回1000/200，200作为困难负样本微调向量模型召回100，召回100的部分作为排序负样本，最终选择20

• 3rd code

• code

• 7th code

• 8th code

• 9th code

• code

---

MRC 信息抽取

kaggle-Tweet Sentiment Extraction

• 1st code discussion

• 2nd discussion

• 3rd code

• 7th code

• base code code

天池-瑞金医院MMC人工智能辅助构建知识图谱

百度aistudio事件抽取比赛

• 12nd code discussion

**CCKS&百度 2019中文短文本的实体链指))

• code

• code

• code

2019之江杯人工智能大赛电商评论观点挖掘赛道

• https://github.com/eguilg/OpinioNet

• https://github.com/srtianxia/opinion_mining

科大讯飞2020事件抽取挑战赛

• https://github.com/WuHuRestaurant/xf_event_extraction2020Top1

• https://github.com/xiaoqian19940510/Event-Extraction

• https://lonepatient.top/2022/07/12/gaiic_2022_ner_top10.html

• https://github.com/luhua-rain/MRC_Competition_Dureader

• https://github.com/YingZiqiang/LES-MMRC-Summary

• https://github.com/basketballandlearn/MRC_Competition_Dureader

---

QA问答

CCKS2019 CKBQA

• 4th code

• 9th code

hinde

• 1st code

• 4th code

• 5th discussion

• 36th discussion

中医文献问题生成挑战

• https://tianchi.aliyun.com/forum/postDetail?spm=5176.12586969.1002.3.2db024ddZShYhb&postId=10854

https://github.com/Dikea/Dialog-System-with-Task-Retrieval-and-Seq2seq https://github.com/xueyouluo/S2S-in-Production

---

其他

kaggle 手语识别

• 2nd discussion