LLM(Large Language Model) Advent Calendar 2024
https://qiita.com/advent-calendar/2024/llm
3日目投稿予定の記事です。
MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
Xiang Yue, Yuansheng Ni, Kai Zhang, Tianyu Zheng, Ruoqi Liu, Ge Zhang, Samuel Stevens, Dongfu Jiang, Weiming Ren, Yuxuan Sun, Cong Wei, Botao Yu, Ruibin Yuan, Renliang Sun, Ming Yin, Boyuan Zheng, Zhenzhu Yang, Yibo Liu, Wenhao Huang, Huan Sun, Yu Su, Wenhu Chen
https://arxiv.org/abs/2311.16502v4
参考文献は、番号が一部抜けています。順次追記します。
単語帳は、一部、単語がうまく分割できず、現在、単語分解するか方法を検討中です。
今しばらくお待ちください。よい方法があれば、コメント欄にご記入いただけると幸いです。
<この項は書きかけです。順次追記します。>
This article is not completed. I will add some words and/or centences in order.
References
[1] Blaise Agu ̈era y Arcas and Peter Norvig. Artificial general intelligence is already here. Noema Magazine, 2023. 1
[2] Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine
Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. Flamingo: a visual language model for few-shot learning. In Advances in Neural Information Processing Systems, 2022. 3, 5
[3] Stanislaw Antol, Aishwarya Agrawal, Jiasen Lu, Margaret Mitchell, Dhruv Batra, C. Lawrence Zitnick, and Devi Parikh. VQA: Visual Question Answering. In International Conference on Computer Vision (ICCV), 2015. 2, 3
[4] AnasAwadalla,IrenaGao,JoshGardner,JackHessel,Yusuf Hanafy, Wanrong Zhu, Kalyani Marathe, Yonatan Bitton, Samir Gadre, Shiori Sagawa, et al. Openflamingo: An open- source framework for training large autoregressive vision- language models. arXiv preprint arXiv:2308.01390, 2023. 3, 5, 6, 15, 16, 17, 18, 19, 20, 21
[5] Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. Qwen-vl: A versatile vision-language model for un- derstanding, localization, text reading, and beyond. arXiv preprint arXiv:2308.12966, 2023. 5, 6, 7, 15, 16, 17, 18, 19, 20, 21
[6] Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, and Sag ̆nak Tas ̧ırlar. Introducing our multimodal models, 2023. 3, 5, 6, 7, 15, 16, 17, 18, 19, 20, 21
[7] Se ́bastienBubeck,VarunChandrasekaran,RonenEldan,Jo- hannes Gehrke, Eric Horvitz, Ece Kamar, Peter Lee, Yin Tat Lee, Yuanzhi Li, Scott Lundberg, et al. Sparks of artificial general intelligence: Early experiments with gpt-4. arXiv preprint arXiv:2303.12712, 2023. 1
[8] Bunny. Bunny-3b. https://github.com/cappuch/ Bunny-Qwen, 2024. GitHub Repository. 15, 16, 17, 18, 19, 20, 21
[9] Xi Chen, Josip Djolonga, Piotr Padlewski, Basil Mustafa, Soravit Changpinyo, Jialin Wu, Carlos Riquelme Ruiz, Se- bastian Goodman, Xiao Wang, Yi Tay, et al. Pali-x: On scaling up a multilingual vision and language model. arXiv preprint arXiv:2305.18565, 2023. 2
[10] Yen-Chun Chen, Linjie Li, Licheng Yu, Ahmed El Kholy, Faisal Ahmed, Zhe Gan, Yu Cheng, and Jingjing Liu. Uniter: Universal image-text representation learning. In European Conference on Computer Vision, pages 104–120, 2020. 3
[11] Zhe Chen, Jiannan Wu, Wenhai Wang, Weijie Su, Guo Chen, Sen Xing, Zhong Muyan, Qinglong Zhang, Xizhou Zhu, Lewei Lu, et al. Internvl: Scaling up vision foundation mod- els and aligning for generic visual-linguistic tasks. arXiv preprint arXiv:2312.14238, 2023. 6, 15, 16, 17, 18, 19, 20, 21
[12] Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhang- hao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yong- hao Zhuang, Joseph E. Gonzalez, Ion Stoica, and Eric P. Xing. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality, 2023. 3, 5, 6, 15, 16, 17, 18, 19, 20, 21
[13] Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:2204.02311, 2022. 1
[14] Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, et al. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416, 2022. 3, 5, 6, 15, 16, 17, 18, 19, 20, 21
[15] Chenhang Cui, Yiyang Zhou, Xinyu Yang, Shirley Wu, Lin- jun Zhang, James Zou, and Huaxiu Yao. Holistic analysis of hallucination in gpt-4v (ision): Bias and interference chal- lenges. arXiv preprint arXiv:2311.03287, 2023. 3, 8
[16] Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, and Steven Hoi. Instructblip: Towards general- purpose vision-language models with instruction tuning. arXiv preprint arXiv:2305.06500, 2023. 2, 3, 5, 6, 7, 15, 16, 17, 18, 19, 20, 21
[17] Xiaoyi Dong, Pan Zhang, Yuhang Zang, Yuhang Cao, Bin Wang, Linke Ouyang, Xilin Wei, Songyang Zhang, Haodong Duan, Maosong Cao, et al. Internlm-xcomposer2: Mastering free-form text-image composition and compre- hension in vision-language large model. arXiv preprint arXiv:2401.16420, 2024. 6, 15, 16, 17, 18, 19, 20, 21
[18] Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. In International Conference on Learning Representa- tions, 2021. 3
[19] Adept Fuyu Team. Adept fuyu-heavy: A new multimodal model. https://www.adept.ai/blog/adept- fuyu-heavy, 2024. 15, 16, 17, 18, 19, 20, 21
[20] Peng Gao, Jiaming Han, Renrui Zhang, Ziyi Lin, Shijie Geng, Aojun Zhou, Wei Zhang, Pan Lu, Conghui He, Xi- angyu Yue, et al. Llama-adapter v2: Parameter-efficient vi- sual instruction model. arXiv preprint arXiv:2304.15010, 2023. 3, 5
[21] Yingqiang Ge, Wenyue Hua, Jianchao Ji, Juntao Tan, Shuyuan Xu, and Yongfeng Zhang. Openagi: When llm meets domain experts. arXiv preprint arXiv:2304.04370, 2023. 1
[22] Google Gemini Team. Gemini: A family of highly capable multimodal models. https : / / storage . googleapis . com / deepmind - media / gemini / gemini_1_report.pdf, 2023. 15, 16, 17, 18, 19, 20, 21, 119
[23] Google Gemini Team. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. https: //storage.googleapis.com/deepmind-media/ gemini/gemini_v1_5_report.pdf, 2024. 6, 15, 119
[24] Yash Goyal, Tejas Khot, Douglas Summers-Stay, Dhruv Ba- tra, and Devi Parikh. Making the v in vqa matter: Elevating
the role of image understanding in visual question answer- ing. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 6904–6913, 2017. 2, 3
[25] Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. Mea- suring massive multitask language understanding. In Inter- national Conference on Learning Representations, 2020. 2
[26] Yupan Huang, Zaiqiao Meng, Fangyu Liu, Yixuan Su, Col- lier Nigel, and Yutong Lu. Sparkles: Unlocking chats across multiple images for multimodal instruction-following mod- els. arXiv preprint arXiv:2308.16463, 2023. 3
[27] Drew A Hudson and Christopher D Manning. Gqa: A new dataset for real-world visual reasoning and compositional question answering. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition, pages 6700–6709, 2019. 3
[28] HyperGAI.Revolutionizingthefuturewithhypergenerative ai. 2024. 15, 16, 17, 18, 19, 20, 21
[29] ChaoJia,YinfeiYang,YeXia,Yi-TingChen,ZaranaParekh, Hieu Pham, Quoc Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig. Scaling up visual and vision-language representa- tion learning with noisy text supervision. In International conference on machine learning, pages 4904–4916. PMLR, 2021. 3
[30] Sahar Kazemzadeh, Vicente Ordonez, Mark Matten, and Tamara Berg. Referitgame: Referring to objects in pho- tographs of natural scenes. In Proceedings of the 2014 con- ference on empirical methods in natural language processing (EMNLP), pages 787–798, 2014. 2
[31] Kunlun. Agi and aigc business skywork. 2024. 15, 16, 17, 18, 19, 20, 21
[32] Ehsan Latif, Gengchen Mai, Matthew Nyaaba, Xuansheng Wu, Ninghao Liu, Guoyu Lu, Sheng Li, Tianming Liu, and Xiaoming Zhai. Artificial general intelligence (agi) for edu- cation. arXiv preprint arXiv:2304.12479, 2023. 1
[33] Bohao Li, Rui Wang, Guangzhi Wang, Yuying Ge, Yix- iao Ge, and Ying Shan. Seed-bench: Benchmarking mul- timodal llms with generative comprehension. arXiv preprint arXiv:2307.16125, 2023. 2, 3
[34] Bo Li, Yuanhan Zhang, Liangyu Chen, Jinghao Wang, Jingkang Yang, and Ziwei Liu. Otter: A multi-modal model with in-context instruction tuning. arXiv preprint arXiv:2305.03726, 2023. 3, 5, 15, 16, 17, 18, 19, 20, 21
[35] Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. Inter- national Conference on Machine Learning, 2023. 2, 3, 5, 6, 7, 15, 16, 17, 18, 19, 20, 21
[36] Lei Li, Yuwei Yin, Shicheng Li, Liang Chen, Peiyi Wang, Shuhuai Ren, Mukai Li, Yazheng Yang, Jingjing Xu, Xu Sun, et al. M3it: A large-scale dataset towards multi- modal multilingual instruction tuning. arXiv preprint arXiv:2306.04387, 2023. 3
[37] Xiujun Li, Xi Yin, Chunyuan Li, Pengchuan Zhang, Xiaowei Hu, Lei Zhang, Lijuan Wang, Houdong Hu, Li Dong, Furu Wei, et al. Oscar: Object-semantics aligned pre-training for vision-language tasks. In Computer Vision–ECCV 2020:
16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXX 16, pages 121–137. Springer, 2020. 3
[38] Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Wayne Xin Zhao, and Ji-Rong Wen. Evaluating object hallucina- tion in large vision-language models. arXiv preprint arXiv:2305.10355, 2023. 3
[39] Ji Lin, Hongxu Yin, Wei Ping, Yao Lu, Pavlo Molchanov, Andrew Tao, Huizi Mao, Jan Kautz, Mohammad Shoeybi, and Song Han. Vila: On pre-training for visual language models. arXiv preprint arXiv:2312.07533, 2023. 6, 15, 16, 17, 18, 19, 20, 21
[40] Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dolla ́r, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, pages 740–755. Springer, 2014. 2, 3
[41] ZiyiLin,ChrisLiu,RenruiZhang,PengGao,LongtianQiu, Han Xiao, Han Qiu, Chen Lin, Wenqi Shao, Keqin Chen, et al. Sphinx: The joint mixing of weights, tasks, and visual embeddings for multi-modal large language models. arXiv preprint arXiv:2311.07575, 2023. 15, 16, 17, 18, 19, 20, 21
[42] Fuxiao Liu, Tianrui Guan, Zongxia Li, Lichang Chen, Yaser Yacoob, Dinesh Manocha, and Tianyi Zhou. Hallusion- bench: You see what you think? or you think what you see? an image-context reasoning benchmark challenging for gpt- 4v (ision), llava-1.5, and other multi-modality models. arXiv preprint arXiv:2310.14566, 2023. 3
[43] Fuxiao Liu, Kevin Lin, Linjie Li, Jianfeng Wang, Yaser Yacoob, and Lijuan Wang. Aligning large multi-modal model with robust instruction tuning. arXiv preprint arXiv:2306.14565, 2023. 3
[44] Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. Improved baselines with visual instruction tuning. arXiv preprint arXiv:2310.03744, 2023. 2, 3, 5, 7, 15, 16, 17, 18, 19, 20, 21
[45] HaotianLiu,ChunyuanLi,QingyangWu,andYongJaeLee. Visual instruction tuning. arXiv preprint arXiv:2304.08485, 2023. 3
[46] Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, and Yong Jae Lee. Llava-next: Im- proved reasoning, ocr, and world knowledge. 2024. 6, 15, 16, 17, 18, 19, 20, 21
[47] YuanLiu,HaodongDuan,YuanhanZhang,BoLi,Songyang Zhang, Wangbo Zhao, Yike Yuan, Jiaqi Wang, Conghui He, Ziwei Liu, et al. Mmbench: Is your multi-modal model an all-around player? arXiv preprint arXiv:2307.06281, 2023. 2, 3
[48] Yuliang Liu, Zhang Li, Hongliang Li, Wenwen Yu, Mingxin Huang, Dezhi Peng, Mingyu Liu, Mingrui Chen, Chunyuan Li, Lianwen Jin, et al. On the hidden mystery of ocr in large multimodal models. arXiv preprint arXiv:2305.07895, 2023. 3
[49] Jiasen Lu, Dhruv Batra, Devi Parikh, and Stefan Lee. Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks. Advances in neural information processing systems, 32, 2019. 3
[50] Pan Lu, Swaroop Mishra, Tanglin Xia, Liang Qiu, Kai-Wei Chang, Song-Chun Zhu, Oyvind Tafjord, Peter Clark, and Ashwin Kalyan. Learn to explain: Multimodal reasoning via thought chains for science question answering. Advances in Neural Information Processing Systems, 35:2507–2521, 2022. 2
[51] Pan Lu, Hritik Bansal, Tony Xia, Jiacheng Liu, Chunyuan Li, Hannaneh Hajishirzi, Hao Cheng, Kai-Wei Chang, Michel Galley, and Jianfeng Gao. Mathvista: Evaluating mathemat- ical reasoning of foundation models in visual contexts. arXiv preprint arXiv:2310.02255, 2023. 3
[52] Kenneth Marino, Mohammad Rastegari, Ali Farhadi, and Roozbeh Mottaghi. Ok-vqa: A visual question answering benchmark requiring external knowledge. In Conference on Computer Vision and Pattern Recognition (CVPR), 2019. 3
[53] Gre ́goireMialon,Cle ́mentineFourrier,CraigSwift,Thomas Wolf, Yann LeCun, and Thomas Scialom. Gaia: a benchmark for general ai assistants. arXiv preprint arXiv:2311.12983, 2023. 1, 3
[54] MiniCPM. Minicpm-v. https://github.com/ OpenBMB/MiniCPM, 2024. GitHub Repository. 15, 16, 17, 18, 19, 20, 21
[55] MiniCPM. Minicpm-v-2, 2024. 15, 16, 17, 18, 19, 20, 21
[56] Masoud Monajatipoor, Liunian Harold Li, Mozhdeh Rouhsedaghat, Lin F Yang, and Kai-Wei Chang. Metavl: Transferring in-context learning ability from language models to vision-language models. arXiv preprint
arXiv:2306.01311, 2023. 3
[57] MeredithRingelMorris,JaschaSohl-dickstein,NoahFiedel,
Tris Warkentin, Allan Dafoe, Aleksandra Faust, Clement Farabet, and Shane Legg. Levels of agi: Opera- tionalizing progress on the path to agi. arXiv preprint arXiv:2311.02462, 2023. 1, 3, 8
[58] OminiLMM. Ominilmm-12b. https://github.com/ OpenBMB/OmniLMM, 2024. GitHub Repository. 15, 16, 17, 18, 19, 20, 21
[59] OpenAI. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023. 1, 6, 15, 16, 17, 18, 19, 20, 21
[60] OpenAI. Gpt-4v(ision) system card, 2023. 2, 6, 7, 15, 16, 17, 18, 19, 20, 21
[61] OpenAI. Gpt-4o. 2024. 6, 15, 119
[62] Aitor Ormazabal, Che Zheng, Cyprien de Masson d’Autume,
Dani Yogatama, Deyu Fu, Donovan Ong, et al. Reka core, flash, and edge: A series of powerful multimodal language models. https://publications.reka.ai/reka- core-tech-report.pdf, 2024. 15, 16, 17, 18, 19, 20, 21, 119
[63] Zhiliang Peng, Wenhui Wang, Li Dong, Yaru Hao, Shaohan Huang, Shuming Ma, and Furu Wei. Kosmos-2: Ground- ing multimodal large language models to the world. arXiv preprint arXiv:2306.14824, 2023. 5, 6, 15, 16, 17, 18, 19, 20, 21
[64] Qwen. Qwen-vl-plus. https://github.com/ QwenLM/Qwen-VL?tab=readme-ov-file#qwen- vl-plus, 2023. GitHub Repository. 15, 16, 17, 18, 19, 20, 21
[65] Qwen. Qwen-vl-max. https : / / github . com / QwenLM/Qwen-VL?tab=readme-ov-file#qwen- vl-max, 2024. GitHub Repository. 6, 15, 16, 17, 18, 19, 20, 21
Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervi- sion. In International conference on machine learning, pages 8748–8763. PMLR, 2021. 3, 5
Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information process- ing systems, 28, 2015. 3
sensenova. Sensechat-vision, 2024. 6, 15, 16, 17, 18, 19, 20, 21
Amanpreet Singh, Vivek Natarjan, Meet Shah, Yu Jiang, Xinlei Chen, Devi Parikh, and Marcus Rohrbach. Towards vqa models that can read. In Proceedings of the IEEE Con- ference on Computer Vision and Pattern Recognition, pages 8317–8326, 2019. 2
Quan Sun, Yufeng Cui, Xiaosong Zhang, Fan Zhang, Qiying Yu, Zhengxiong Luo, Yueze Wang, Yongming Rao, Jingjing Liu, Tiejun Huang, et al. Generative multimodal models are in-context learners. arXiv preprint arXiv:2312.13286, 2023. 15, 16, 17, 18, 19, 20, 21
Hao Tan and Mohit Bansal. Lxmert: Learning cross- modality encoder representations from transformers. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP- IJCNLP), pages 5100–5111, 2019. 3
Claude Team. Introducing the next generation of claude.
https://www.anthropic.com/news/claude-3- family, 2024. 6, 15, 119
InfiMM Team. Infimm: Advancing multimodal understand- ing from flamingo’s legacy through diverse llm integration, 2024. 15, 16, 17, 18, 19, 20, 21
Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothe ́e Lacroix, Baptiste Rozie`re, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023. 1, 5
Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288, 2023. 6, 15, 16, 17, 18, 19, 20, 21
Junyang Wang, Yiyang Zhou, Guohai Xu, Pengcheng Shi, Chenlin Zhao, Haiyang Xu, Qinghao Ye, Ming Yan, Ji Zhang, Jihua Zhu, et al. Evaluation and analysis of hal- lucination in large vision-language models. arXiv preprint arXiv:2308.15126, 2023. 3
Weihan Wang, Qingsong Lv, Wenmeng Yu, Wenyi Hong, Ji Qi, Yan Wang, Junhui Ji, Zhuoyi Yang, Lei Zhao, Xixuan Song, et al. Cogvlm: Visual expert for pretrained language models. arXiv preprint arXiv:2311.03079, 2023. 2, 5, 6, 15, 16, 17, 18, 19, 20, 21
[78] Zirui Wang, Jiahui Yu, Adams Wei Yu, Zihang Dai, Yulia
Tsvetkov, and Yuan Cao. Simvlm: Simple visual language model pretraining with weak supervision. In International Conference on Learning Representations, 2021. 3
[79] Peng Xu, Wenqi Shao, Kaipeng Zhang, Peng Gao, Shuo Liu, Meng Lei, Fanqing Meng, Siyuan Huang, Yu Qiao, and Ping Luo. Lvlm-ehub: A comprehensive evaluation benchmark for large vision-language models. arXiv preprint arXiv:2306.09265, 2023. 3
[80] Zhengyuan Yang, Linjie Li, Kevin Lin, Jianfeng Wang, Chung-Ching Lin, Zicheng Liu, and Lijuan Wang. The dawn of lmms: Preliminary explorations with gpt-4v (ision). arXiv preprint arXiv:2309.17421, 2023. 2
[81] Qinghao Ye, Haiyang Xu, Guohai Xu, Jiabo Ye, Ming Yan, Yiyang Zhou, Junyang Wang, Anwen Hu, Pengcheng Shi, Yaya Shi, et al. mplug-owl: Modularization empowers large language models with multimodality. arXiv preprint arXiv:2304.14178, 2023. 3
[82] Qinghao Ye, Haiyang Xu, Jiabo Ye, Ming Yan, Haowei Liu, Qi Qian, Ji Zhang, Fei Huang, and Jingren Zhou. mplug-owl2: Revolutionizing multi-modal large language model with modality collaboration. arXiv preprint arXiv:2311.04257, 2023. 3, 5, 15, 16, 17, 18, 19, 20, 21
[83] Zhenfei Yin, Jiong Wang, Jianjian Cao, Zhelun Shi, Dingn- ing Liu, Mukai Li, Lu Sheng, Lei Bai, Xiaoshui Huang, Zhiyong Wang, et al. Lamm: Language-assisted multi- modal instruction-tuning dataset, framework, and bench- mark. arXiv preprint arXiv:2306.06687, 2023. 2, 3
[84] AlexYoung,BeiChen,ChaoLi,ChengenHuang,GeZhang, Guanwei Zhang, Heng Li, Jiangcheng Zhu, Jianqun Chen, Jing Chang, et al. Yi: Open foundation models by 01. ai. arXiv preprint arXiv:2403.04652, 2024. 6, 15, 16, 17, 18, 19, 20, 21
[85] Jiahui Yu, Zirui Wang, Vijay Vasudevan, Legg Yeung, Mo- jtaba Seyedhosseini, and Yonghui Wu. Coca: Contrastive captioners are image-text foundation models. TMLR, 2022. 3
[86] Weihao Yu, Zhengyuan Yang, Linjie Li, Jianfeng Wang, Kevin Lin, Zicheng Liu, Xinchao Wang, and Lijuan Wang. Mm-vet: Evaluating large multimodal models for integrated capabilities. arXiv preprint arXiv:2308.02490, 2023. 2, 3
[87] Pengchuan Zhang, Xiujun Li, Xiaowei Hu, Jianwei Yang, Lei Zhang, Lijuan Wang, Yejin Choi, and Jianfeng Gao. Vinvl: Revisiting visual representations in vision-language models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5579–5588, 2021. 3
[88] Renrui Zhang, Jiaming Han, Aojun Zhou, Xiangfei Hu, Shilin Yan, Pan Lu, Hongsheng Li, Peng Gao, and Yu Qiao. Llama-adapter: Efficient fine-tuning of language models with zero-init attention. arXiv preprint arXiv:2303.16199, 2023. 3, 6, 15, 16, 17, 18, 19, 20, 21
[89] Bo Zhao, Boya Wu, and Tiejun Huang. Svit: Scaling up visual instruction tuning. arXiv preprint arXiv:2307.04087, 2023. 3, 15, 16, 17, 18, 19, 20, 21
term list
| no | wprd | count |
|---|---|---|
| 1 | the | 523 |
| 2 | and | 459 |
| 3 | of | 455 |
| 4 | a | 442 |
| 5 | to | 379 |
| 6 | b | 320 |
| 7 | image | 277 |
| 8 | in | 258 |
| 9 | v | 195 |
| 10 | back | 172 |
| 11 | error | 168 |
| 12 | c | 164 |
| 13 | d | 145 |
| 14 | gpt | 132 |
| 15 | models | 125 |
| 16 | table | 125 |
| 17 | figure | 124 |
| 18 | is | 119 |
| 19 | arxiv | 101 |
| 20 | case | 101 |
| 21 | for | 101 |
| 22 | figures | 96 |
| 23 | on | 95 |
| 24 | sample | 94 |
| 25 | subfield | 93 |
| 26 | index | 88 |
| 27 | list | 87 |
| 28 | groundtruth | 86 |
| 29 | option | 86 |
| 30 | vl | 82 |
| 31 | t | 81 |
| 32 | as | 80 |
| 33 | we | 80 |
| 34 | with | 76 |
| 35 | reasoning | 74 |
| 36 | perceptual | 66 |
| 37 | s | 64 |
| 38 | category | 62 |
| 39 | mmmu | 62 |
| 40 | from | 59 |
| 41 | m | 57 |
| 42 | text | 57 |
| 43 | language | 56 |
| 44 | questions | 56 |
| 45 | expert | 54 |
| 46 | art | 52 |
| 47 | multimodal | 52 |
| 48 | correct | 51 |
| 49 | e | 51 |
| 50 | visual | 51 |
| 51 | knowledge | 50 |
| 52 | preprint | 50 |
| 53 | that | 48 |
| 54 | however | 46 |
| 55 | results | 46 |
| 56 | are | 45 |
| 57 | data | 45 |
| 58 | errorcategory | 45 |
| 59 | design | 44 |
| 60 | li | 44 |
| 61 | this | 44 |
| 62 | vision | 43 |
| 63 | by | 42 |
| 64 | llava | 42 |
| 65 | model | 42 |
| 66 | engineering | 41 |
| 67 | it | 41 |
| 68 | or | 41 |
| 69 | science | 40 |
| 70 | so | 40 |
| 71 | medicine | 39 |
| 72 | open | 38 |
| 73 | correctcase | 37 |
| 74 | benchmark | 36 |
| 75 | errorerrorreason | 36 |
| 76 | large | 36 |
| 77 | qwen | 36 |
| 78 | be | 34 |
| 79 | chat | 34 |
| 80 | history | 34 |
| 81 | wang | 34 |
| 82 | an | 33 |
| 83 | choice | 33 |
| 84 | different | 33 |
| 85 | lack | 33 |
| 86 | question | 33 |
| 87 | zhang | 33 |
| 88 | h | 32 |
| 89 | lmms | 32 |
| 90 | flan | 31 |
| 91 | xxl | 31 |
| 92 | chemistry | 30 |
| 93 | performance | 30 |
| 94 | types | 30 |
| 95 | r | 29 |
| 96 | these | 29 |
| 97 | best | 28 |
| 98 | gemini | 28 |
| 99 | images | 27 |
| 100 | liu | 27 |
| 101 | test | 27 |
| 102 | validation | 27 |
| 103 | perceptualerror | 26 |
| 104 | therefore | 26 |
| 105 | understanding | 26 |
| 106 | where | 26 |
| 107 | al | 25 |
| 108 | et | 25 |
| 109 | fuyu | 25 |
| 110 | ocr | 25 |
| 111 | shot | 25 |
| 112 | subjects | 25 |
| 113 | each | 24 |
| 114 | k | 24 |
| 115 | level | 24 |
| 116 | llama | 24 |
| 117 | music | 24 |
| 118 | only | 24 |
| 119 | blip | 23 |
| 120 | health | 23 |
| 121 | such | 23 |
| 122 | clinical | 22 |
| 123 | input | 22 |
| 124 | overall | 22 |
| 125 | reka | 22 |
| 126 | set | 22 |
| 127 | source | 22 |
| 128 | but | 21 |
| 129 | chen | 21 |
| 130 | our | 21 |
| 131 | vicuna | 21 |
| 132 | yi | 21 |
| 133 | agi | 20 |
| 134 | biology | 20 |
| 135 | both | 20 |
| 136 | computer | 20 |
| 137 | geography | 20 |
| 138 | has | 20 |
| 139 | instructblip | 20 |
| 140 | instruction | 20 |
| 141 | minicpm | 20 |
| 142 | mm | 20 |
| 143 | provided | 20 |
| 144 | subject | 20 |
| 145 | authors | 19 |
| 146 | benchmarks | 19 |
| 147 | can | 19 |
| 148 | conference | 19 |
| 149 | i | 19 |
| 150 | learning | 19 |
| 151 | llms | 19 |
| 152 | psychology | 19 |
| 153 | tasks | 19 |
| 154 | theory | 19 |
| 155 | answer | 18 |
| 156 | like | 18 |
| 157 | materials | 18 |
| 158 | multi | 18 |
| 159 | multiple | 18 |
| 160 | plus | 18 |
| 161 | should | 18 |
| 162 | which | 18 |
| 163 | appendix | 17 |
| 164 | caption | 17 |
| 165 | errors | 17 |
| 166 | internvl | 17 |
| 167 | math | 17 |
| 168 | th | 17 |
| 169 | thus | 17 |
| 170 | x | 17 |
| 171 | yu | 17 |
| 172 | adept | 16 |
| 173 | context | 16 |
| 174 | disciplines | 16 |
| 175 | economics | 16 |
| 176 | given | 16 |
| 177 | have | 16 |
| 178 | literature | 16 |
| 179 | more | 16 |
| 180 | paper | 16 |
| 181 | rad | 16 |
| 182 | reasoningerror | 16 |
| 183 | social | 16 |
| 184 | agriculture | 15 |
| 185 | all | 15 |
| 186 | analysis | 15 |
| 187 | college | 15 |
| 188 | dataset | 15 |
| 189 | human | 15 |
| 190 | its | 15 |
| 191 | let | 15 |
| 192 | lin | 15 |
| 193 | marco | 15 |
| 194 | openflamingo | 15 |
| 195 | other | 15 |
| 196 | physiology | 15 |
| 197 | specific | 15 |
| 198 | u | 15 |
| 199 | while | 15 |
| 200 | also | 14 |
| 201 | at | 14 |
| 202 | business | 14 |
| 203 | difficulty | 14 |
| 204 | domain | 14 |
| 205 | evaluation | 14 |
| 206 | foundation | 14 |
| 207 | information | 14 |
| 208 | ision | 14 |
| 209 | ln | 14 |
| 210 | not | 14 |
| 211 | otter | 14 |
| 212 | p | 14 |
| 213 | pro | 14 |
| 214 | their | 14 |
| 215 | tuning | 14 |
| 216 | xl | 14 |
| 217 | accounting | 13 |
| 218 | across | 13 |
| 219 | added | 13 |
| 220 | annotators | 13 |
| 221 | arrow | 13 |
| 222 | basic | 13 |
| 223 | between | 13 |
| 224 | challenges | 13 |
| 225 | electronics | 13 |
| 226 | github | 13 |
| 227 | hpt | 13 |
| 228 | https | 13 |
| 229 | management | 13 |
| 230 | marketing | 13 |
| 231 | nation | 13 |
| 232 | now | 13 |
| 233 | q | 13 |
| 234 | research | 13 |
| 235 | second | 13 |
| 236 | thecorrectansweris | 13 |
| 237 | there | 13 |
| 238 | towards | 13 |
| 239 | adapter | 12 |
| 240 | existing | 12 |
| 241 | few | 12 |
| 242 | finance | 12 |
| 243 | general | 12 |
| 244 | knowledgeerrorreason | 12 |
| 245 | lackofknowledge | 12 |
| 246 | lu | 12 |
| 247 | medium | 12 |
| 248 | norepinephrine | 12 |
| 249 | options | 12 |
| 250 | owl | 12 |
| 251 | pages | 12 |
| 252 | pathology | 12 |
| 253 | pharmacy | 12 |
| 254 | processing | 12 |
| 255 | random | 12 |
| 256 | shows | 12 |
| 257 | significant | 12 |
| 258 | sociology | 12 |
| 259 | st | 12 |
| 260 | tech | 12 |
| 261 | various | 12 |
| 262 | vqa | 12 |
| 263 | will | 12 |
| 264 | yang | 12 |
| 265 | annotation | 11 |
| 266 | based | 11 |
| 267 | been | 11 |
| 268 | bunny | 11 |
| 269 | challenging | 11 |
| 270 | cogvlm | 11 |
| 271 | crepe | 11 |
| 272 | diagnostics | 11 |
| 273 | diverse | 11 |
| 274 | frequent | 11 |
| 275 | heavy | 11 |
| 276 | humanities | 11 |
| 277 | kosmos | 11 |
| 278 | lens | 11 |
| 279 | max | 11 |
| 280 | mechanical | 11 |
| 281 | mechanics | 11 |
| 282 | medical | 11 |
| 283 | minigpt | 11 |
| 284 | mplug | 11 |
| 285 | no | 11 |
| 286 | report | 11 |
| 287 | subfields | 11 |
| 288 | total | 11 |
| 289 | training | 11 |
| 290 | type | 11 |
| 291 | ultra | 11 |
| 292 | wei | 11 |
| 293 | would | 11 |
| 294 | zhou | 11 |
| 295 | advanced | 10 |
| 296 | ai | 10 |
| 297 | animal | 10 |
| 298 | ca | 10 |
| 299 | cl | 10 |
| 300 | core | 10 |
| 301 | f | 10 |
| 302 | fine | 10 |
| 303 | flash | 10 |
| 304 | huang | 10 |
| 305 | infimm | 10 |
| 306 | internlm | 10 |
| 307 | laboratory | 10 |
| 308 | light | 10 |
| 309 | manage | 10 |
| 310 | modal | 10 |
| 311 | perception | 10 |
| 312 | performing | 10 |
| 313 | physics | 10 |
| 314 | pressure | 10 |
| 315 | problems | 10 |
| 316 | quality | 10 |
| 317 | representation | 10 |
| 318 | sensechat | 10 |
| 319 | sphinx | 10 |
| 320 | study | 10 |
| 321 | svit | 10 |
| 322 | team | 10 |
| 323 | textbooks | 10 |
| 324 | up | 10 |
| 325 | vila | 10 |
| 326 | xcomposer | 10 |
| 327 | abilities | 9 |
| 328 | annotated | 9 |
| 329 | architecture | 9 |
| 330 | author | 9 |
| 331 | claude | 9 |
| 332 | com | 9 |
| 333 | complex | 9 |
| 334 | control | 9 |
| 335 | example | 9 |
| 336 | experts | 9 |
| 337 | financial | 9 |
| 338 | first | 9 |
| 339 | if | 9 |
| 340 | interleaved | 9 |
| 341 | international | 9 |
| 342 | landsat | 9 |
| 343 | omnilmm | 9 |
| 344 | physical | 9 |
| 345 | point | 9 |
| 346 | preview | 9 |
| 347 | proceedings | 9 |
| 348 | process | 9 |
| 349 | project | 9 |
| 350 | public | 9 |
| 351 | recognition | 9 |
| 352 | see | 9 |
| 353 | selected | 9 |
| 354 | shares | 9 |
| 355 | sharesat | 9 |
| 356 | skywork | 9 |
| 357 | step | 9 |
| 358 | still | 9 |
| 359 | structure | 9 |
| 360 | sun | 9 |
| 361 | textual | 9 |
| 362 | through | 9 |
| 363 | tyramine | 9 |
| 364 | without | 9 |
| 365 | world | 9 |
| 366 | xu | 9 |
| 367 | you | 9 |
| 368 | yue | 9 |
| 369 | achieve | 8 |
| 370 | additionally | 8 |
| 371 | advantage | 8 |
| 372 | answers | 8 |
| 373 | any | 8 |
| 374 | bold | 8 |
| 375 | capabilities | 8 |
| 376 | cm | 8 |
| 377 | collection | 8 |
| 378 | comparative | 8 |
| 379 | contributed | 8 |
| 380 | cost | 8 |
| 381 | depth | 8 |
| 382 | detailed | 8 |
| 383 | diagrams | 8 |
| 384 | discipline | 8 |
| 385 | easy | 8 |
| 386 | edge | 8 |
| 387 | emu | 8 |
| 388 | evaluating | 8 |
| 389 | format | 8 |
| 390 | g | 8 |
| 391 | gao | 8 |
| 392 | historyquestion | 8 |
| 393 | including | 8 |
| 394 | intelligence | 8 |
| 395 | into | 8 |
| 396 | json | 8 |
| 397 | kai | 8 |
| 398 | kj | 8 |
| 399 | most | 8 |
| 400 | nh | 8 |
| 401 | one | 8 |
| 402 | peng | 8 |
| 403 | plant | 8 |
| 404 | playground | 8 |
| 405 | power | 8 |
| 406 | representations | 8 |
| 407 | significantly | 8 |
| 408 | systems | 8 |
| 409 | underlined | 8 |
| 410 | when | 8 |
| 411 | worst | 8 |
| 412 | wu | 8 |
| 413 | xiang | 8 |
| 414 | zephyr | 8 |
| 415 | zhao | 8 |
| 416 | accuracy | 7 |
| 417 | air | 7 |
| 418 | airbase | 7 |
| 419 | approximately | 7 |
| 420 | aqua | 7 |
| 421 | artificial | 7 |
| 422 | computerscience | 7 |
| 423 | distribution | 7 |
| 424 | drug | 7 |
| 425 | due | 7 |
| 426 | effect | 7 |
| 427 | energy | 7 |
| 428 | explanation | 7 |
| 429 | further | 7 |
| 430 | inorganic | 7 |
| 431 | ji | 7 |
| 432 | lei | 7 |
| 433 | less | 7 |
| 434 | market | 7 |
| 435 | mmhg | 7 |
| 436 | natural | 7 |
| 437 | next | 7 |
| 438 | pe | 7 |
| 439 | pre | 7 |
| 440 | production | 7 |
| 441 | response | 7 |
| 442 | se | 7 |
| 443 | sources | 7 |
| 444 | su | 7 |
| 445 | theansweris | 7 |
| 446 | transaction | 7 |
| 447 | understand | 7 |
| 448 | university | 7 |
| 449 | veh | 7 |
| 450 | very | 7 |
| 451 | y | 7 |
| 452 | year | 7 |
| 453 | zheng | 7 |
| 454 | among | 6 |
| 455 | bd | 6 |
| 456 | breadth | 6 |
| 457 | broad | 6 |
| 458 | captioning | 6 |
| 459 | cause | 6 |
| 460 | ch | 6 |
| 461 | chang | 6 |
| 462 | chemical | 6 |
| 463 | chunyuan | 6 |
| 464 | comprehensive | 6 |
| 465 | consider | 6 |
| 466 | direction | 6 |
| 467 | ended | 6 |
| 468 | european | 6 |
| 469 | evaluate | 6 |
| 470 | examples | 6 |
| 471 | file | 6 |
| 472 | fixed | 6 |
| 473 | flamingo | 6 |
| 474 | fluid | 6 |
| 475 | following | 6 |
| 476 | genetics | 6 |
| 477 | han | 6 |
| 478 | hard | 6 |
| 479 | hu | 6 |
| 480 | interpretation | 6 |
| 481 | j | 6 |
| 482 | jianfeng | 6 |
| 483 | key | 6 |
| 484 | leading | 6 |
| 485 | lee | 6 |
| 486 | massive | 6 |
| 487 | mathematical | 6 |
| 488 | mc | 6 |
| 489 | minimal | 6 |
| 490 | n | 6 |
| 491 | ni | 6 |
| 492 | oc | 6 |
| 493 | organic | 6 |
| 494 | over | 6 |
| 495 | painting | 6 |
| 496 | pattern | 6 |
| 497 | perform | 6 |
| 498 | relatively | 6 |
| 499 | scaling | 6 |
| 500 | terra | 6 |
| 501 | thescaleofthephotographsis | 6 |
| 502 | trade | 6 |
| 503 | two | 6 |
| 504 | use | 6 |
| 505 | weget | 6 |
| 506 | well | 6 |
| 507 | were | 6 |
| 508 | what | 6 |
| 509 | ye | 6 |
| 510 | yin | 6 |
| 511 | zhu | 6 |
| 512 | adults | 5 |
| 513 | advances | 5 |
| 514 | after | 5 |
| 515 | answering | 5 |
| 516 | available | 5 |
| 517 | baselines | 5 |
| 518 | because | 5 |
| 519 | besides | 5 |
| 520 | blood | 5 |
| 521 | calculus | 5 |
| 522 | carbon | 5 |
| 523 | charts | 5 |
| 524 | clinicalmedicine | 5 |
| 525 | collected | 5 |
| 526 | collecting | 5 |
| 527 | common | 5 |
| 528 | conducted | 5 |
| 529 | cover | 5 |
| 530 | cpu | 5 |
| 531 | critical | 5 |
| 532 | curation | 5 |
| 533 | current | 5 |
| 534 | designed | 5 |
| 535 | double | 5 |
| 536 | dynamics | 5 |
| 537 | efficient | 5 |
| 538 | epidemiology | 5 |
| 539 | experiments | 5 |
| 540 | fields | 5 |
| 541 | find | 5 |
| 542 | focus | 5 |
| 543 | forstate | 5 |
| 544 | fp | 5 |
| 545 | ge | 5 |
| 546 | graph | 5 |
| 547 | here | 5 |
| 548 | hox | 5 |
| 549 | improvement | 5 |
| 550 | instances | 5 |
| 551 | instead | 5 |
| 552 | levels | 5 |
| 553 | lijuan | 5 |
| 554 | ma | 5 |
| 555 | made | 5 |
| 556 | main | 5 |
| 557 | major | 5 |
| 558 | mi | 5 |
| 559 | microbiology | 5 |
| 560 | might | 5 |
| 561 | modality | 5 |
| 562 | nature | 5 |
| 563 | often | 5 |
| 564 | paintings | 5 |
| 565 | pan | 5 |
| 566 | part | 5 |
| 567 | per | 5 |
| 568 | pharmacology | 5 |
| 569 | phenelzine | 5 |
| 570 | photographs | 5 |
| 571 | principle | 5 |
| 572 | progress | 5 |
| 573 | protocol | 5 |
| 574 | range | 5 |
| 575 | reject | 5 |
| 576 | repository | 5 |
| 577 | require | 5 |
| 578 | requires | 5 |
| 579 | root | 5 |
| 580 | scale | 5 |
| 581 | secondary | 5 |
| 582 | section | 5 |
| 583 | sheng | 5 |
| 584 | signal | 5 |
| 585 | since | 5 |
| 586 | single | 5 |
| 587 | six | 5 |
| 588 | skilled | 5 |
| 589 | state | 5 |
| 590 | statistics | 5 |
| 591 | students | 5 |
| 592 | synthesis | 5 |
| 593 | system | 5 |
| 594 | tables | 5 |
| 595 | three | 5 |
| 596 | units | 5 |
| 597 | updated | 5 |
| 598 | within | 5 |
| 599 | yan | 5 |
| 600 | yes | 5 |
| 601 | ab | 4 |
| 602 | absolute | 4 |
| 603 | accurate | 4 |
| 604 | anatomy | 4 |
| 605 | ande | 4 |
| 606 | approach | 4 |
| 607 | arts | 4 |
| 608 | assess | 4 |
| 609 | atom | 4 |
| 610 | average | 4 |
| 611 | axis | 4 |
| 612 | bandc | 4 |
| 613 | barrel | 4 |
| 614 | batra | 4 |
| 615 | biochemistry | 4 |
| 616 | block | 4 |
| 617 | bo | 4 |
| 618 | bothtimeseriesaremeanstationary | 4 |
| 619 | breakdown | 4 |
| 620 | calculations | 4 |
| 621 | cao | 4 |
| 622 | cardiovascular | 4 |
| 623 | change | 4 |
| 624 | chemistryquestion | 4 |
| 625 | civil | 4 |
| 626 | co | 4 |
| 627 | cocaine | 4 |
| 628 | comparison | 4 |
| 629 | compiler | 4 |
| 630 | complexity | 4 |
| 631 | considerations | 4 |
| 632 | contamination | 4 |
| 633 | contemporary | 4 |
| 634 | contribution | 4 |
| 635 | converging | 4 |
| 636 | copyright | 4 |
| 637 | corporate | 4 |
| 638 | correctly | 4 |
| 639 | covering | 4 |
| 640 | dec | 4 |
| 641 | deliberate | 4 |
| 642 | derive | 4 |
| 643 | devi | 4 |
| 644 | diagnosticsandlabmedicine | 4 |
| 645 | directlabor | 4 |
| 646 | eliminate | 4 |
| 647 | embedding | 4 |
| 648 | end | 4 |
| 649 | eng | 4 |
| 650 | eric | 4 |
| 651 | even | 4 |
| 652 | exams | 4 |
| 653 | finishes | 4 |
| 654 | follow | 4 |
| 655 | foot | 4 |
| 656 | force | 4 |
| 657 | fpr | 4 |
| 658 | gap | 4 |
| 659 | generation | 4 |
| 660 | geometric | 4 |
| 661 | hallucination | 4 |
| 662 | handling | 4 |
| 663 | hao | 4 |
| 664 | heterogeneous | 4 |
| 665 | highly | 4 |
| 666 | huan | 4 |
| 667 | identify | 4 |
| 668 | ieee | 4 |
| 669 | indexvalueatt | 4 |
| 670 | introduces | 4 |
| 671 | joint | 4 |
| 672 | kevin | 4 |
| 673 | law | 4 |
| 674 | lead | 4 |
| 675 | liang | 4 |
| 676 | linjie | 4 |
| 677 | llm | 4 |
| 678 | long | 4 |
| 679 | machine | 4 |
| 680 | macroeconomics | 4 |
| 681 | mapping | 4 |
| 682 | mask | 4 |
| 683 | means | 4 |
| 684 | meng | 4 |
| 685 | ming | 4 |
| 686 | minor | 4 |
| 687 | mri | 4 |
| 688 | must | 4 |
| 689 | necessary | 4 |
| 690 | neural | 4 |
| 691 | neuropathology | 4 |
| 692 | new | 4 |
| 693 | non | 4 |
| 694 | number | 4 |
| 695 | o | 4 |
| 696 | objects | 4 |
| 697 | ofhand | 4 |
| 698 | openai | 4 |
| 699 | optical | 4 |
| 700 | outputs | 4 |
| 701 | parallel | 4 |
| 702 | parikh | 4 |
| 703 | path | 4 |
| 704 | perfect | 4 |
| 705 | personyears | 4 |
| 706 | peter | 4 |
| 707 | pharmaceutical | 4 |
| 708 | photo | 4 |
| 709 | physiologyquestion | 4 |
| 710 | present | 4 |
| 711 | presents | 4 |
| 712 | pretraining | 4 |
| 713 | probability | 4 |
| 714 | prompt | 4 |
| 715 | provide | 4 |
| 716 | qwenvl | 4 |
| 717 | radiology | 4 |
| 718 | regions | 4 |
| 719 | resources | 4 |
| 720 | role | 4 |
| 721 | samples | 4 |
| 722 | sci | 4 |
| 723 | selection | 4 |
| 724 | shi | 4 |
| 725 | show | 4 |
| 726 | shown | 4 |
| 727 | skills | 4 |
| 728 | solution | 4 |
| 729 | some | 4 |
| 730 | song | 4 |
| 731 | specialized | 4 |
| 732 | stage | 4 |
| 733 | starts | 4 |
| 734 | statement | 4 |
| 735 | states | 4 |
| 736 | strong | 4 |
| 737 | structures | 4 |
| 738 | surveying | 4 |
| 739 | tail | 4 |
| 740 | than | 4 |
| 741 | thecorrectoptionis | 4 |
| 742 | thedouble | 4 |
| 743 | thefatheriscomparedtoagypsy | 4 |
| 744 | thegeographicextentofthemonetizationofeurasianeconomies | 4 |
| 745 | theorderisfromlefttoright | 4 |
| 746 | thermodynamics | 4 |
| 747 | thespeedofaishalfthatofb | 4 |
| 748 | they | 4 |
| 749 | tianyu | 4 |
| 750 | tn | 4 |
| 751 | toptobottom | 4 |
| 752 | tuned | 4 |
| 753 | using | 4 |
| 754 | vaultedroofing | 4 |
| 755 | velocity | 4 |
| 756 | vet | 4 |
| 757 | wenhu | 4 |
| 758 | whichofthefollowingisthemostlikelydiagnosis | 4 |
| 759 | whorl | 4 |
| 760 | work | 4 |
| 761 | yuan | 4 |
| 762 | yuansheng | 4 |
| 763 | zero | 4 |
| 764 | ability | 3 |
| 765 | achieves | 3 |
| 766 | add | 3 |
| 767 | addition | 3 |
| 768 | adhere | 3 |
| 769 | adherence | 3 |
| 770 | administration | 3 |
| 771 | adopted | 3 |
| 772 | adult | 3 |
| 773 | aligning | 3 |
| 774 | andc | 3 |
| 775 | andstate | 3 |
| 776 | aob | 3 |
| 777 | aorta | 3 |
| 778 | architectureandengineering | 3 |
| 779 | area | 3 |
| 780 | areproductionzonesofprovenoilreserves | 3 |
| 781 | around | 3 |
| 782 | arttheory | 3 |
| 783 | aspect | 3 |
| 784 | atestoftwoindependentmeans | 3 |
| 785 | att | 3 |
| 786 | attention | 3 |
| 787 | au | 3 |
| 788 | augmented | 3 |
| 789 | bai | 3 |
| 790 | beam | 3 |
| 791 | being | 3 |
| 792 | believe | 3 |
| 793 | better | 3 |
| 794 | beyond | 3 |
| 795 | biological | 3 |
| 796 | bod | 3 |
| 797 | body | 3 |
| 798 | bothmovetostate | 3 |
| 799 | buttheydifferoninput | 3 |
| 800 | capability | 3 |
| 801 | capable | 3 |
| 802 | carefully | 3 |
| 803 | categories | 3 |
| 804 | cbt | 3 |
| 805 | cell | 3 |
| 806 | changes | 3 |
| 807 | chao | 3 |
| 808 | character | 3 |
| 809 | child | 3 |
| 810 | choosingthematchingterm | 3 |
| 811 | chung | 3 |
| 812 | classical | 3 |
| 813 | clef | 3 |
| 814 | clip | 3 |
| 815 | cod | 3 |
| 816 | comics | 3 |
| 817 | conclusion | 3 |
| 818 | consistent | 3 |
| 819 | contain | 3 |
| 820 | contains | 3 |
| 821 | contributions | 3 |
| 822 | crepes | 3 |
| 823 | criticism | 3 |
| 824 | ctr | 3 |
| 825 | cui | 3 |
| 826 | dandy | 3 |
| 827 | deep | 3 |
| 828 | demonstrate | 3 |
| 829 | details | 3 |
| 830 | determine | 3 |
| 831 | develop | 3 |
| 832 | development | 3 |
| 833 | dhruv | 3 |
| 834 | diagram | 3 |
| 835 | directmaterials | 3 |
| 836 | discussions | 3 |
| 837 | disease | 3 |
| 838 | do | 3 |
| 839 | dong | 3 |
| 840 | drama | 3 |
| 841 | drawing | 3 |
| 842 | duan | 3 |
| 843 | econometrics | 3 |
| 844 | effects | 3 |
| 845 | egoism | 3 |
| 846 | electrocardiography | 3 |
| 847 | elements | 3 |
| 848 | empirical | 3 |
| 849 | employed | 3 |
| 850 | encoder | 3 |
| 851 | engineeringquestion | 3 |
| 852 | ensure | 3 |
| 853 | epidemiologyquestion | 3 |
| 854 | essential | 3 |
| 855 | ethical | 3 |
| 856 | evident | 3 |
| 857 | evolution | 3 |
| 858 | exchange | 3 |
| 859 | expertise | 3 |
| 860 | expressions | 3 |
| 861 | fails | 3 |
| 862 | falsepositives | 3 |
| 863 | family | 3 |
| 864 | features | 3 |
| 865 | findings | 3 |
| 866 | formats | 3 |
| 867 | four | 3 |
| 868 | fromthetable | 3 |
| 869 | fundamental | 3 |
| 870 | future | 3 |
| 871 | generative | 3 |
| 872 | genes | 3 |
| 873 | geographyquestion | 3 |
| 874 | geometry | 3 |
| 875 | go | 3 |
| 876 | goal | 3 |
| 877 | gqa | 3 |
| 878 | graphic | 3 |
| 879 | grounding | 3 |
| 880 | haiyang | 3 |
| 881 | haotian | 3 |
| 882 | he | 3 |
| 883 | head | 3 |
| 884 | heart | 3 |
| 885 | height | 3 |
| 886 | higher | 3 |
| 887 | how | 3 |
| 888 | iii | 3 |
| 889 | illustrated | 3 |
| 890 | imperialist | 3 |
| 891 | importance | 3 |
| 892 | improved | 3 |
| 893 | incorrect | 3 |
| 894 | increase | 3 |
| 895 | indicating | 3 |
| 896 | industrial | 3 |
| 897 | inputs | 3 |
| 898 | insights | 3 |
| 899 | insteadof | 3 |
| 900 | instructions | 3 |
| 901 | introduce | 3 |
| 902 | investment | 3 |
| 903 | involve | 3 |
| 904 | isms | 3 |
| 905 | itislikelytobesn | 3 |
| 906 | jae | 3 |
| 907 | jingjing | 3 |
| 908 | jun | 3 |
| 909 | junyang | 3 |
| 910 | justinian | 3 |
| 911 | kpa | 3 |
| 912 | layer | 3 |
| 913 | led | 3 |
| 914 | length | 3 |
| 915 | licensing | 3 |
| 916 | limitations | 3 |
| 917 | line | 3 |
| 918 | linear | 3 |
| 919 | literaturequestion | 3 |
| 920 | lower | 3 |
| 921 | making | 3 |
| 922 | many | 3 |
| 923 | maoi | 3 |
| 924 | maturenewborn | 3 |
| 925 | may | 3 |
| 926 | measure | 3 |
| 927 | mechanicalengineering | 3 |
| 928 | meet | 3 |
| 929 | meticulously | 3 |
| 930 | mmbench | 3 |
| 931 | modalities | 3 |
| 932 | modern | 3 |
| 933 | months | 3 |
| 934 | musicquestion | 3 |
| 935 | mutant | 3 |
| 936 | name | 3 |
| 937 | naming | 3 |
| 938 | need | 3 |
| 939 | needs | 3 |
| 940 | netincome | 3 |
| 941 | notable | 3 |
| 942 | note | 3 |
| 943 | object | 3 |
| 944 | oninput | 3 |
| 945 | online | 3 |
| 946 | ophthalmic | 3 |
| 947 | opportunity | 3 |
| 948 | optics | 3 |
| 949 | opus | 3 |
| 950 | out | 3 |
| 951 | overhead | 3 |
| 952 | overheadrate | 3 |
| 953 | oxygen | 3 |
| 954 | pathologyquestion | 3 |
| 955 | 3 | |
| 956 | phentolamine | 3 |
| 957 | photography | 3 |
| 958 | possible | 3 |
| 959 | posterior | 3 |
| 960 | prior | 3 |
| 961 | proprietary | 3 |
| 962 | publichealth | 3 |
| 963 | qinghao | 3 |
| 964 | qiu | 3 |
| 965 | queries | 3 |
| 966 | quizzes | 3 |
| 967 | rather | 3 |
| 968 | rays | 3 |
| 969 | redistribution | 3 |
| 970 | reduction | 3 |
| 971 | reference | 3 |
| 972 | region | 3 |
| 973 | regulations | 3 |
| 974 | relatedhand | 3 |
| 975 | ren | 3 |
| 976 | renrui | 3 |
| 977 | represents | 3 |
| 978 | respiratory | 3 |
| 979 | responses | 3 |
| 980 | room | 3 |
| 981 | rule | 3 |
| 982 | ruoqi | 3 |
| 983 | rupturedberryaneurysm | 3 |
| 984 | savior | 3 |
| 985 | scans | 3 |
| 986 | scienceqa | 3 |
| 987 | sciences | 3 |
| 988 | seed | 3 |
| 989 | seen | 3 |
| 990 | sense | 3 |
| 991 | separate | 3 |
| 992 | shapes | 3 |
| 993 | sheets | 3 |
| 994 | short | 3 |
| 995 | shortpastern | 3 |
| 996 | solvefor | 3 |
| 997 | space | 3 |
| 998 | standard | 3 |
| 999 | starting | 3 |
| 1000 | startsandfinisheswithoutanyinterleaving | 3 |
| 1001 | stateerror | 3 |
| 1002 | steven | 3 |
| 1003 | strategic | 3 |
| 1004 | structural | 3 |
| 1005 | substantial | 3 |
| 1006 | suggests | 3 |
| 1007 | sulfur | 3 |
| 1008 | supervision | 3 |
| 1009 | surgery | 3 |
| 1010 | symptom | 3 |
| 1011 | tallbacksofchairsandlampsatthecornersofdiningtables | 3 |
| 1012 | tan | 3 |
| 1013 | task | 3 |
| 1014 | theconfigurationatc | 3 |
| 1015 | them | 3 |
| 1016 | then | 3 |
| 1017 | thepainting | 3 |
| 1018 | thepontomedullaryjunction | 3 |
| 1019 | theyarenotequivalent | 3 |
| 1020 | third | 3 |
| 1021 | thomas | 3 |
| 1022 | those | 3 |
| 1023 | top | 3 |
| 1024 | totaldirectlabordollars | 3 |
| 1025 | totalfactoryoverhead | 3 |
| 1026 | typically | 3 |
| 1027 | typo | 3 |
| 1028 | united | 3 |
| 1029 | usingtheequation | 3 |
| 1030 | visionmayberestoredwithconcavelensandrefractivesurgery | 3 |
| 1031 | walkersyndrome | 3 |
| 1032 | way | 3 |
| 1033 | weighted | 3 |
| 1034 | wide | 3 |
| 1035 | writing | 3 |
| 1036 | xi | 3 |
| 1037 | xia | 3 |
| 1038 | xiao | 3 |
| 1039 | yiyang | 3 |
| 1040 | yong | 3 |
| 1041 | yuanhan | 3 |
| 1042 | _ | 2 |
| 1043 | _report | 2 |
| 1044 | aandb | 2 |
| 1045 | abarbicanandbattlements | 2 |
| 1046 | able | 2 |
| 1047 | above | 2 |
| 1048 | abstract | 2 |
| 1049 | accepting | 2 |
| 1050 | accuracies | 2 |
| 1051 | accurately | 2 |
| 1052 | address | 2 |
| 1053 | adjacent | 2 |
| 1054 | adrenergic | 2 |
| 1055 | advancements | 2 |
| 1056 | adversarial | 2 |
| 1057 | advertisements | 2 |
| 1058 | advice | 2 |
| 1059 | agieval | 2 |
| 1060 | ahigherrooftomakeupfortheshortcolumns | 2 |
| 1061 | ahmed | 2 |
| 1062 | aim | 2 |
| 1063 | aims | 2 |
| 1064 | algebra | 2 |
| 1065 | algorithm | 2 |
| 1066 | align | 2 |
| 1067 | along | 2 |
| 1068 | alpha | 2 |
| 1069 | ambiguities | 2 |
| 1070 | amoatandcrenellations | 2 |
| 1071 | amothertellshersontostopwhining | 2 |
| 1072 | amount | 2 |
| 1073 | analog | 2 |
| 1074 | analyzed | 2 |
| 1075 | anda | 2 |
| 1076 | andcpu | 2 |
| 1077 | anddemocraticgovernments | 2 |
| 1078 | andm | 2 |
| 1079 | andmouthdiseaseintheplacebogroup | 2 |
| 1080 | andn | 2 |
| 1081 | andrequiresfurtherwork | 2 |
| 1082 | andsoon | 2 |
| 1083 | andthenthefirstonemightresume | 2 |
| 1084 | andthere | 2 |
| 1085 | andthisremainsunchanged | 2 |
| 1086 | aneurysm | 2 |
| 1087 | angle | 2 |
| 1088 | annotations | 2 |
| 1089 | another | 2 |
| 1090 | answererrorreason | 2 |
| 1091 | antagonist | 2 |
| 1092 | anterior | 2 |
| 1093 | anxietydisorder | 2 |
| 1094 | aojun | 2 |
| 1095 | apply | 2 |
| 1096 | areas | 2 |
| 1097 | artsquestion | 2 |
| 1098 | asaresult | 2 |
| 1099 | ascending | 2 |
| 1100 | assistedinsitukeratomileusis | 2 |
| 1101 | associated | 2 |
| 1102 | attachedgroupsandtheiratomicnumbers | 2 |
| 1103 | avoid | 2 |
| 1104 | avoided | 2 |
| 1105 | bansal | 2 |
| 1106 | baptiste | 2 |
| 1107 | bar | 2 |
| 1108 | basedmethod | 2 |
| 1109 | basedontheimageprovided | 2 |
| 1110 | basicmedicalscience | 2 |
| 1111 | beginningretainedearnings | 2 |
| 1112 | benchmarking | 2 |
| 1113 | beta | 2 |
| 1114 | betweensphere | 2 |
| 1115 | bias | 2 |
| 1116 | biodiversity | 2 |
| 1117 | biostatistics | 2 |
| 1118 | blocked | 2 |
| 1119 | blocks | 2 |
| 1120 | bone | 2 |
| 1121 | botany | 2 |
| 1122 | boyuan | 2 |
| 1123 | breast | 2 |
| 1124 | bridge | 2 |
| 1125 | brownstemrot | 2 |
| 1126 | butitfailedtocorrectlymaptheidstothecorrespondingillustrationsinthefigure | 2 |
| 1127 | cal | 2 |
| 1128 | calculatethemanufacturingcostperunitforproducta | 2 |
| 1129 | calculatethetotalmanufacturingcostforproducta | 2 |
| 1130 | calculatethetotaloverheadrate | 2 |
| 1131 | calculatethework | 2 |
| 1132 | calculation | 2 |
| 1133 | calculusquestion | 2 |
| 1134 | candd | 2 |
| 1135 | cartoon | 2 |
| 1136 | cartoons | 2 |
| 1137 | cases | 2 |
| 1138 | categorize | 2 |
| 1139 | cbtismoreeffectivethannotreatmentandmoreeffectivethanmeditation | 2 |
| 1140 | cbtisnotaseffectiveasmeditation | 2 |
| 1141 | cd | 2 |
| 1142 | chains | 2 |
| 1143 | challenge | 2 |
| 1144 | challenginghimtoconsiderthemultitudeofinterpretationsthepaintingrepresents | 2 |
| 1145 | chaotic | 2 |
| 1146 | cheng | 2 |
| 1147 | chi | 2 |
| 1148 | children | 2 |
| 1149 | chlorine | 2 |
| 1150 | chris | 2 |
| 1151 | chun | 2 |
| 1152 | cini | 2 |
| 1153 | circuit | 2 |
| 1154 | circulatory | 2 |
| 1155 | clark | 2 |
| 1156 | clc | 2 |
| 1157 | clearly | 2 |
| 1158 | close | 2 |
| 1159 | closed | 2 |
| 1160 | coauthors | 2 |
| 1161 | coca | 2 |
| 1162 | collaboration | 2 |
| 1163 | collect | 2 |
| 1164 | collectiveeffervescence | 2 |
| 1165 | commonsense | 2 |
| 1166 | completes | 2 |
| 1167 | comprehension | 2 |
| 1168 | comprising | 2 |
| 1169 | concave | 2 |
| 1170 | conceived | 2 |
| 1171 | conceptualization | 2 |
| 1172 | conghui | 2 |
| 1173 | considered | 2 |
| 1174 | consistency | 2 |
| 1175 | consistently | 2 |
| 1176 | consumerismandnationalidentities | 2 |
| 1177 | contrast | 2 |
| 1178 | converge | 2 |
| 1179 | copying | 2 |
| 1180 | corresponding | 2 |
| 1181 | correspondstoregionssuchasnortherncanadaandpartsofrussia | 2 |
| 1182 | cotton | 2 |
| 1183 | could | 2 |
| 1184 | covers | 2 |
| 1185 | creating | 2 |
| 1186 | cross | 2 |
| 1187 | crucial | 2 |
| 1188 | ct | 2 |
| 1189 | cu | 2 |
| 1190 | cvf | 2 |
| 1191 | dai | 2 |
| 1192 | daily | 2 |
| 1193 | datasets | 2 |
| 1194 | dawn | 2 |
| 1195 | decorativerhythmandrepetition | 2 |
| 1196 | decrease | 2 |
| 1197 | deeply | 2 |
| 1198 | deepmind | 2 |
| 1199 | default | 2 |
| 1200 | definition | 2 |
| 1201 | degradation | 2 |
| 1202 | degrees | 2 |
| 1203 | dehghani | 2 |
| 1204 | demonstrates | 2 |
| 1205 | dental | 2 |
| 1206 | depositionequilibrium | 2 |
| 1207 | designquestion | 2 |
| 1208 | designs | 2 |
| 1209 | despair | 2 |
| 1210 | detection | 2 |
| 1211 | determinethechangeininternalenergy | 2 |
| 1212 | deterministicfiniteautomaton | 2 |
| 1213 | dev | 2 |
| 1214 | developed | 2 |
| 1215 | diastolic | 2 |
| 1216 | difficult | 2 |
| 1217 | difficulties | 2 |
| 1218 | digital | 2 |
| 1219 | diminished | 2 |
| 1220 | direct | 2 |
| 1221 | disparity | 2 |
| 1222 | disred | 2 |
| 1223 | distributions | 2 |
| 1224 | dividends | 2 |
| 1225 | documents | 2 |
| 1226 | doesn | 2 |
| 1227 | doing | 2 |
| 1228 | domainis | 2 |
| 1229 | domains | 2 |
| 1230 | dongxu | 2 |
| 1231 | du | 2 |
| 1232 | duetothelackofspecificknowledgeabout | 2 |
| 1233 | dynamicsquestion | 2 |
| 1234 | eachofmass | 2 |
| 1235 | easyquestion | 2 |
| 1236 | eccv | 2 |
| 1237 | ecology | 2 |
| 1238 | effectively | 2 |
| 1239 | ehub | 2 |
| 1240 | eisblue | 2 |
| 1241 | electrical | 2 |
| 1242 | electromagnetism | 2 |
| 1243 | elementary | 2 |
| 1244 | embeddings | 2 |
| 1245 | energyandpower | 2 |
| 1246 | enhanced | 2 |
| 1247 | enhancements | 2 |
| 1248 | enhancing | 2 |
| 1249 | equilibrium | 2 |
| 1250 | ers | 2 |
| 1251 | etc | 2 |
| 1252 | evalai | 2 |
| 1253 | exhibit | 2 |
| 1254 | exists | 2 |
| 1255 | explanations | 2 |
| 1256 | explicitly | 2 |
| 1257 | expressed | 2 |
| 1258 | extensive | 2 |
| 1259 | external | 2 |
| 1260 | extraction | 2 |
| 1261 | faisal | 2 |
| 1262 | falls | 2 |
| 1263 | faster | 2 |
| 1264 | fe | 2 |
| 1265 | fiction | 2 |
| 1266 | finding | 2 |
| 1267 | firstly | 2 |
| 1268 | flawed | 2 |
| 1269 | focal | 2 |
| 1270 | focallength | 2 |
| 1271 | follows | 2 |
| 1272 | forc | 2 |
| 1273 | forchoice | 2 |
| 1274 | forexample | 2 |
| 1275 | foribssuffererswithoutananxietydisorder | 2 |
| 1276 | formula | 2 |
| 1277 | forsphere | 2 |
| 1278 | found | 2 |
| 1279 | framework | 2 |
| 1280 | free | 2 |
| 1281 | fromthegivenimage | 2 |
| 1282 | fromwhich | 2 |
| 1283 | frozen | 2 |
| 1284 | furu | 2 |
| 1285 | fuxiao | 2 |
| 1286 | gaia | 2 |
| 1287 | gas | 2 |
| 1288 | gene | 2 |
| 1289 | geneinamousewasreplacedwithahox | 2 |
| 1290 | geneticsquestion | 2 |
| 1291 | geotechnical | 2 |
| 1292 | give | 2 |
| 1293 | giventhis | 2 |
| 1294 | gives | 2 |
| 1295 | 2 | |
| 1296 | googleapis | 2 |
| 1297 | goyal | 2 |
| 1298 | gpa | 2 |
| 1299 | groundheight | 2 |
| 1300 | group | 2 |
| 1301 | guardcells | 2 |
| 1302 | guo | 2 |
| 1303 | guohai | 2 |
| 1304 | haiku | 2 |
| 1305 | handpart | 2 |
| 1306 | haodong | 2 |
| 1307 | hardsubject | 2 |
| 1308 | heat | 2 |
| 1309 | help | 2 |
| 1310 | helping | 2 |
| 1311 | hence | 2 |
| 1312 | hg | 2 |
| 1313 | hidden | 2 |
| 1314 | hierarchical | 2 |
| 1315 | hierarchicalscale | 2 |
| 1316 | highlight | 2 |
| 1317 | highlights | 2 |
| 1318 | hoi | 2 |
| 1319 | holistic | 2 |
| 1320 | horizontal | 2 |
| 1321 | horror | 2 |
| 1322 | houdong | 2 |
| 1323 | hugo | 2 |
| 1324 | humanpapillomavirusinfection | 2 |
| 1325 | humans | 2 |
| 1326 | hydrogen | 2 |
| 1327 | hypertensive | 2 |
| 1328 | hyung | 2 |
| 1329 | ican | 2 |
| 1330 | ifahox | 2 |
| 1331 | immediacy | 2 |
| 1332 | immunology | 2 |
| 1333 | improve | 2 |
| 1334 | inaccuracies | 2 |
| 1335 | incidencedensity | 2 |
| 1336 | include | 2 |
| 1337 | included | 2 |
| 1338 | includes | 2 |
| 1339 | incorrectly | 2 |
| 1340 | increases | 2 |
| 1341 | indeed | 2 |
| 1342 | indicate | 2 |
| 1343 | indicates | 2 |
| 1344 | initial | 2 |
| 1345 | instance | 2 |
| 1346 | intercept | 2 |
| 1347 | interface | 2 |
| 1348 | interleavedprocessingoccurswhentwotransactionsareprocessedalternately | 2 |
| 1349 | interleaving | 2 |
| 1350 | internet | 2 |
| 1351 | interpret | 2 |
| 1352 | interval | 2 |
| 1353 | inthepoliticalcartoon | 2 |
| 1354 | inthepradomuseuminmadrid | 2 |
| 1355 | inthesecondimage | 2 |
| 1356 | inthestudyofkingphilipiv | 2 |
| 1357 | intricate | 2 |
| 1358 | introducing | 2 |
| 1359 | introduction | 2 |
| 1360 | involving | 2 |
| 1361 | io | 2 |
| 1362 | ipit | 2 |
| 1363 | ipitdisplaysastrongseasonality | 2 |
| 1364 | isminimal | 2 |
| 1365 | istheaccelerationduetogravity | 2 |
| 1366 | italy | 2 |
| 1367 | jack | 2 |
| 1368 | jacob | 2 |
| 1369 | james | 2 |
| 1370 | jiabo | 2 |
| 1371 | jiahui | 2 |
| 1372 | jiaming | 2 |
| 1373 | jiang | 2 |
| 1374 | jiasen | 2 |
| 1375 | jingren | 2 |
| 1376 | jointly | 2 |
| 1377 | junnan | 2 |
| 1378 | kgak | 2 |
| 1379 | kln | 2 |
| 1380 | labor | 2 |
| 1381 | lamm | 2 |
| 1382 | laser | 2 |
| 1383 | later | 2 |
| 1384 | lawrence | 2 |
| 1385 | le | 2 |
| 1386 | leaderboard | 2 |
| 1387 | leads | 2 |
| 1388 | leaving | 2 |
| 1389 | left | 2 |
| 1390 | leftventricle | 2 |
| 1391 | legg | 2 |
| 1392 | lewis | 2 |
| 1393 | limit | 2 |
| 1394 | limited | 2 |
| 1395 | linguistic | 2 |
| 1396 | llfollowthesesteps | 2 |
| 1397 | lmm | 2 |
| 1398 | log | 2 |
| 1399 | logic | 2 |
| 1400 | lone | 2 |
| 1401 | longer | 2 |
| 1402 | longpasternbone | 2 |
| 1403 | lookatthesituationinthe | 2 |
| 1404 | low | 2 |
| 1405 | luo | 2 |
| 1406 | lvlm | 2 |
| 1407 | lxmert | 2 |
| 1408 | macroeconomicsquestion | 2 |
| 1409 | magnitude | 2 |
| 1410 | managerial | 2 |
| 1411 | manuscript | 2 |
| 1412 | mao | 2 |
| 1413 | maps | 2 |
| 1414 | marks | 2 |
| 1415 | mathvista | 2 |
| 1416 | meaningthatonestarts | 2 |
| 1417 | meanwhile | 2 |
| 1418 | measuring | 2 |
| 1419 | mechanicsquestion | 2 |
| 1420 | meconiumaspirationsyndrome | 2 |
| 1421 | media | 2 |
| 1422 | medicinal | 2 |
| 1423 | mediumsubject | 2 |
| 1424 | meeting | 2 |
| 1425 | metavl | 2 |
| 1426 | methods | 2 |
| 1427 | mfroma | 2 |
| 1428 | mg | 2 |
| 1429 | middle | 2 |
| 1430 | mishra | 2 |
| 1431 | mitralregurgitation | 2 |
| 1432 | mmicl | 2 |
| 1433 | mmocr | 2 |
| 1434 | modernhistory | 2 |
| 1435 | module | 2 |
| 1436 | mohammad | 2 |
| 1437 | monochromatic | 2 |
| 1438 | moreover | 2 |
| 1439 | morris | 2 |
| 1440 | mostafa | 2 |
| 1441 | moving | 2 |
| 1442 | mukai | 2 |
| 1443 | multilingual | 2 |
| 1444 | multimodality | 2 |
| 1445 | nano | 2 |
| 1446 | narrative | 2 |
| 1447 | nationxhascomparativeadvantageinpaperproductionandshouldtradepapertonationyinexchangeforcrepes | 2 |
| 1448 | nationygivesupproducing | 2 |
| 1449 | network | 2 |
| 1450 | neurosciences | 2 |
| 1451 | nogpt | 2 |
| 1452 | noneoftheotheranswers | 2 |
| 1453 | normally | 2 |
| 1454 | notsure | 2 |
| 1455 | nuclear | 2 |
| 1456 | occur | 2 |
| 1457 | offered | 2 |
| 1458 | ohm | 2 |
| 1459 | ok | 2 |
| 1460 | ominilmm | 2 |
| 1461 | once | 2 |
| 1462 | ones | 2 |
| 1463 | openbmb | 2 |
| 1464 | operating | 2 |
| 1465 | opportunitycostof | 2 |
| 1466 | oppressor | 2 |
| 1467 | optic | 2 |
| 1468 | oscar | 2 |
| 1469 | others | 2 |
| 1470 | ov | 2 |
| 1471 | overview | 2 |
| 1472 | pairs | 2 |
| 1473 | panningblur | 2 |
| 1474 | parameter | 2 |
| 1475 | participated | 2 |
| 1476 | pathophysiology | 2 |
| 1477 | patientswithnon | 2 |
| 1478 | patterns | 2 |
| 1479 | pengcheng | 2 |
| 1480 | pengchuan | 2 |
| 1481 | percentile | 2 |
| 1482 | personality | 2 |
| 1483 | peutz | 2 |
| 1484 | phalanx | 2 |
| 1485 | photos | 2 |
| 1486 | photoscale | 2 |
| 1487 | phrases | 2 |
| 1488 | ping | 2 |
| 1489 | piotr | 2 |
| 1490 | pivotal | 2 |
| 1491 | played | 2 |
| 1492 | plots | 2 |
| 1493 | pmlr | 2 |
| 1494 | poetry | 2 |
| 1495 | pointe | 2 |
| 1496 | pointf | 2 |
| 1497 | posed | 2 |
| 1498 | poses | 2 |
| 1499 | potential | 2 |
| 1500 | presence | 2 |
| 1501 | price | 2 |
| 1502 | primarily | 2 |
| 1503 | primary | 2 |
| 1504 | principles | 2 |
| 1505 | priorityorder | 2 |
| 1506 | privacy | 2 |
| 1507 | processingquestion | 2 |
| 1508 | producing | 2 |
| 1509 | producta | 2 |
| 1510 | productasalesquantity | 2 |
| 1511 | productb | 2 |
| 1512 | productc | 2 |
| 1513 | prohibit | 2 |
| 1514 | projects | 2 |
| 1515 | pronounced | 2 |
| 1516 | propranolol | 2 |
| 1517 | prosperity | 2 |
| 1518 | providing | 2 |
| 1519 | psychologyquestion | 2 |
| 1520 | ptosisalready | 2 |
| 1521 | purpose | 2 |
| 1522 | puts | 2 |
| 1523 | pwave | 2 |
| 1524 | qi | 2 |
| 1525 | qiao | 2 |
| 1526 | qrscomplex | 2 |
| 1527 | qwenlm | 2 |
| 1528 | ran | 2 |
| 1529 | rapid | 2 |
| 1530 | rate | 2 |
| 1531 | rateofreturn | 2 |
| 1532 | ratio | 2 |
| 1533 | ray | 2 |
| 1534 | reach | 2 |
| 1535 | readme | 2 |
| 1536 | real | 2 |
| 1537 | reason | 2 |
| 1538 | recalling | 2 |
| 1539 | recent | 2 |
| 1540 | recently | 2 |
| 1541 | receptors | 2 |
| 1542 | referring | 2 |
| 1543 | refine | 2 |
| 1544 | refractive | 2 |
| 1545 | regular | 2 |
| 1546 | rejecttoanswer | 2 |
| 1547 | rekacore | 2 |
| 1548 | release | 2 |
| 1549 | relevant | 2 |
| 1550 | religion | 2 |
| 1551 | remedy | 2 |
| 1552 | reported | 2 |
| 1553 | represent | 2 |
| 1554 | represented | 2 |
| 1555 | representing | 2 |
| 1556 | repurposed | 2 |
| 1557 | requiring | 2 |
| 1558 | researchquestion | 2 |
| 1559 | respectively | 2 |
| 1560 | result | 2 |
| 1561 | retainedearningstobereported | 2 |
| 1562 | review | 2 |
| 1563 | revolutionizing | 2 |
| 1564 | right | 2 |
| 1565 | rightventricle | 2 |
| 1566 | rigorous | 2 |
| 1567 | robust | 2 |
| 1568 | robustness | 2 |
| 1569 | roomwithinaroom | 2 |
| 1570 | round | 2 |
| 1571 | sa | 2 |
| 1572 | sampled | 2 |
| 1573 | savingthemfrompovertyoroppressionandbringingthemtrade | 2 |
| 1574 | scene | 2 |
| 1575 | scenes | 2 |
| 1576 | scope | 2 |
| 1577 | sculpture | 2 |
| 1578 | sebastian | 2 |
| 1579 | select | 2 |
| 1580 | selective | 2 |
| 1581 | selects | 2 |
| 1582 | sequence | 2 |
| 1583 | several | 2 |
| 1584 | shade | 2 |
| 1585 | shall | 2 |
| 1586 | shao | 2 |
| 1587 | share | 2 |
| 1588 | sheet | 2 |
| 1589 | shen | 2 |
| 1590 | shijie | 2 |
| 1591 | shortcomings | 2 |
| 1592 | showingawillingnesstobecomparedtogreatspanishpaintersofthepast | 2 |
| 1593 | shuai | 2 |
| 1594 | simpedanceinthes | 2 |
| 1595 | simple | 2 |
| 1596 | simvlm | 2 |
| 1597 | sites | 2 |
| 1598 | siyuan | 2 |
| 1599 | size | 2 |
| 1600 | sketches | 2 |
| 1601 | sleg | 2 |
| 1602 | socialsci | 2 |
| 1603 | solid | 2 |
| 1604 | songyang | 2 |
| 1605 | sonnet | 2 |
| 1606 | sparkles | 2 |
| 1607 | speciesbdescendedfromspeciesa | 2 |
| 1608 | specifically | 2 |
| 1609 | springer | 2 |
| 1610 | square | 2 |
| 1611 | standardized | 2 |
| 1612 | standards | 2 |
| 1613 | startsafterafinishesandcompleteswithoutbeinginterleavedwithanyothertransaction | 2 |
| 1614 | statistical | 2 |
| 1615 | stem | 2 |
| 1616 | steps | 2 |
| 1617 | storage | 2 |
| 1618 | stored | 2 |
| 1619 | student | 2 |
| 1620 | subarachnoidspace | 2 |
| 1621 | success | 2 |
| 1622 | sufficient | 2 |
| 1623 | sustained | 2 |
| 1624 | synthesisquestion | 2 |
| 1625 | tab | 2 |
| 1626 | tackle | 2 |
| 1627 | tasked | 2 |
| 1628 | taxonomy | 2 |
| 1629 | tay | 2 |
| 1630 | technical | 2 |
| 1631 | technique | 2 |
| 1632 | testing | 2 |
| 1633 | textualunderstandingerror | 2 |
| 1634 | theartist | 2 |
| 1635 | thecabinisdepressurizedandtheoxygenmaskfallsfromtheceiling | 2 |
| 1636 | thecorrectcalculationshouldbe | 2 |
| 1637 | thedangeroflettinggoofadream | 2 |
| 1638 | thediffusionofculturaltraditionsalongeurasiantraderoutes | 2 |
| 1639 | theentranceoflightandairintothehall | 2 |
| 1640 | theextenttowhichgovernmenteconomicpoliciesineurasiaintheperiod | 2 |
| 1641 | theincidencedensity | 2 |
| 1642 | theinequitiesofsocieties | 2 |
| 1643 | thejamdensity | 2 |
| 1644 | themodel | 2 |
| 1645 | themostlikelydiagnosisis | 2 |
| 1646 | themousemaydevelopnoheadandtwotails | 2 |
| 1647 | themousemaydeveloptwoheadsandnotail | 2 |
| 1648 | thentheotherstartsbeforethefirstonefinishes | 2 |
| 1649 | theorderis | 2 |
| 1650 | theoryquestion | 2 |
| 1651 | thepatientisapost | 2 |
| 1652 | theperspectiveofthecartoonististhattheunitedstateshasbeenasaviortothenationsbroughtunderitscontrol | 2 |
| 1653 | theregionboundedbythegraphasshownabove | 2 |
| 1654 | thesearethecaseswherebothtestsarepositive | 2 |
| 1655 | thesewomenwanttheirchildrentobeeducated | 2 |
| 1656 | thespreadoftechnologicalinnovationsacrossregionsineurasia | 2 |
| 1657 | thetypeofalkylsubstituentbpresent | 2 |
| 1658 | thetypeofheterocyclicringcpresent | 2 |
| 1659 | thetypeofsubstituentaonthearomaticring | 2 |
| 1660 | theunitedstatesisseenasfulfillingwhichofthefollowingroles | 2 |
| 1661 | thevalueoftheindexis | 2 |
| 1662 | think | 2 |
| 1663 | thisconditionoftenoccursinelderlypeople | 2 |
| 1664 | thisisincorrect | 2 |
| 1665 | thisphenomenoncannotbefixedbylasik | 2 |
| 1666 | thisquestioncallsforknowledgerelatedtothestimulusmaterial | 2 |
| 1667 | thisstatementappearstobetrue | 2 |
| 1668 | thistumormayrepresentthemostcommontypeofintraocularneoplasm | 2 |
| 1669 | tiejun | 2 |
| 1670 | timelines | 2 |
| 1671 | tofind | 2 |
| 1672 | tofindthesteady | 2 |
| 1673 | took | 2 |
| 1674 | tool | 2 |
| 1675 | tot | 2 |
| 1676 | totalmanufacturingcostforproducta | 2 |
| 1677 | touvron | 2 |
| 1678 | train | 2 |
| 1679 | trainable | 2 |
| 1680 | trained | 2 |
| 1681 | transactionaoncpu | 2 |
| 1682 | transactionboncpu | 2 |
| 1683 | transformers | 2 |
| 1684 | trend | 2 |
| 1685 | truenegatives | 2 |
| 1686 | underscore | 2 |
| 1687 | underscores | 2 |
| 1688 | unit | 2 |
| 1689 | uniter | 2 |
| 1690 | universal | 2 |
| 1691 | unknown | 2 |
| 1692 | unlocking | 2 |
| 1693 | uponinspection | 2 |
| 1694 | uptodistinguishitfrommelanoma | 2 |
| 1695 | url | 2 |
| 1696 | used | 2 |
| 1697 | va | 2 |
| 1698 | variable | 2 |
| 1699 | vaultedroof | 2 |
| 1700 | vb | 2 |
| 1701 | vce | 2 |
| 1702 | vdoesn | 2 |
| 1703 | ventriculardepolarization | 2 |
| 1704 | version | 2 |
| 1705 | vfailstointerprettheimage | 2 |
| 1706 | vilbert | 2 |
| 1707 | vinvl | 2 |
| 1708 | visionmayberestoredwithconvexlensandrefractivesurgery | 2 |
| 1709 | visually | 2 |
| 1710 | vit | 2 |
| 1711 | vrecalledtherightknowledgeandmadetherightreasoning | 2 |
| 1712 | w | 2 |
| 1713 | weak | 2 |
| 1714 | web | 2 |
| 1715 | wecan | 2 |
| 1716 | wecandeduce | 2 |
| 1717 | weightedindexofthethreestocksforthefirstperiod | 2 |
| 1718 | weightedindexvalueat | 2 |
| 1719 | wenqi | 2 |
| 1720 | whatisthemostlikelydiagnosis | 2 |
| 1721 | whichisincorrect | 2 |
| 1722 | whichisnotexplicitlymarkedinthefigurebutisonlydescribedintext | 2 |
| 1723 | whichofthesepicturesshowsthereconciliationofegoismandother | 2 |
| 1724 | willbeactivatedandinhibittheseedlingtripleresponse | 2 |
| 1725 | withthehpointingtothebackground | 2 |
| 1726 | withtheswitchinposition | 2 |
| 1727 | won | 2 |
| 1728 | wouldn | 2 |
| 1729 | www | 2 |
| 1730 | xiaowei | 2 |
| 1731 | 2 | |
| 1732 | xiujun | 2 |
| 1733 | yacoob | 2 |
| 1734 | yao | 2 |
| 1735 | yaser | 2 |
| 1736 | years | 2 |
| 1737 | yifan | 2 |
| 1738 | ying | 2 |
| 1739 | youaretravelingonaplanewithasmallchild | 2 |
| 1740 | yuhang | 2 |
| 1741 | yuheng | 2 |
| 1742 | zhai | 2 |
| 1743 | zhe | 2 |
| 1744 | zhengyuan | 2 |
| 1745 | zhong | 2 |
| 1746 | zhuang | 2 |
| 1747 | zicheng | 2 |
| 1748 | zirui | 2 |
| 1749 | zitnick | 2 |
| 1750 | ziwei | 2 |
| 1751 | ziyi | 2 |
| 1752 | zou | 2 |
| 合計 | 5,688 | 20,849 |
合計は出現数１の単語を含みます。