MAP-Neo: Highly Capable and Transparent Bilingual Large Language Model Series.
Ge ZhangScott QuJiaheng LiuChenchen ZhangChenghua LinChou Leuang YuDanny PanEsther ChengJie LiuQunshu LinRaven YuanTuney ZhengWei PangXinrun DuYiming LiangYinghao MaYizhi LiZiyang MaBill Y. LinEmmanouil BenetosHuan YangJunting ZhouKaijing MaMinghao LiuMorry NiuNoah WangQuehry QueRuibo LiuSine LiuShawn GuoSoren GaoWangchunshu ZhouXinyue ZhangYizhi ZhouYubo WangYuelin BaiYuhan ZhangYuxiang ZhangZenith WangZhenzhu YangZijian ZhaoJiajun ZhangWanli OuyangWenhao HuangWenhu ChenPublished in: CoRR (2024)
Keyphrases
- language model
- language modeling
- cross language retrieval
- cross lingual
- multiword
- translation model
- n gram
- probabilistic model
- document retrieval
- test collection
- statistical machine translation
- query expansion
- speech recognition
- information retrieval
- retrieval model
- language modelling
- machine translation
- query terms
- mixture model
- statistical language models
- context sensitive
- vector space model
- document ranking
- smoothing methods
- maximum a posteriori
- ad hoc information retrieval
- chinese english
- cross language
- pseudo relevance feedback
- retrieval effectiveness
- language models for information retrieval
- language model for information retrieval
- query translation
- hidden markov models
- clustering algorithm