no code implementations • 22 Apr 2024 • Marah Abdin, Sam Ade Jacobs, Ammar Ahmad Awan, Jyoti Aneja, Ahmed Awadallah, Hany Awadalla, Nguyen Bach, Amit Bahree, Arash Bakhtiari, Harkirat Behl, Alon Benhaim, Misha Bilenko, Johan Bjorck, Sébastien Bubeck, Martin Cai, Caio César Teodoro Mendes, Weizhu Chen, Vishrav Chaudhary, Parul Chopra, Allie Del Giorno, Gustavo de Rosa, Matthew Dixon, Ronen Eldan, Dan Iter, Amit Garg, Abhishek Goswami, Suriya Gunasekar, Emman Haider, Junheng Hao, Russell J. Hewett, Jamie Huynh, Mojan Javaheripi, Xin Jin, Piero Kauffmann, Nikos Karampatziakis, Dongwoo Kim, Mahoud Khademi, Lev Kurilenko, James R. Lee, Yin Tat Lee, Yuanzhi Li, Chen Liang, Weishung Liu, Eric Lin, Zeqi Lin, Piyush Madan, Arindam Mitra, Hardik Modi, Anh Nguyen, Brandon Norick, Barun Patra, Daniel Perez-Becker, Thomas Portet, Reid Pryzant, Heyang Qin, Marko Radmilac, Corby Rosset, Sambudha Roy, Olatunji Ruwase, Olli Saarikivi, Amin Saied, Adil Salim, Michael Santacroce, Shital Shah, Ning Shang, Hiteshi Sharma, Xia Song, Masahiro Tanaka, Xin Wang, Rachel Ward, Guanhua Wang, Philipp Witte, Michael Wyatt, Can Xu, Jiahang Xu, Sonali Yadav, Fan Yang, ZiYi Yang, Donghan Yu, Chengruidong Zhang, Cyril Zhang, Jianwen Zhang, Li Lyna Zhang, Yi Zhang, Yue Zhang, Yunan Zhang, Xiren Zhou
We introduce phi-3-mini, a 3. 8 billion parameter language model trained on 3. 3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3. 5 (e. g., phi-3-mini achieves 69% on MMLU and 8. 38 on MT-bench), despite being small enough to be deployed on a phone.
no code implementations • 4 Apr 2024 • Corby Rosset, Ching-An Cheng, Arindam Mitra, Michael Santacroce, Ahmed Awadallah, Tengyang Xie
In this paper, we introduce Direct Nash Optimization (DNO), a provable and scalable algorithm that marries the simplicity and stability of contrastive learning with theoretical generality from optimizing general preferences.
no code implementations • 27 Feb 2024 • Corby Rosset, Ho-Lam Chung, Guanghui Qin, Ethan C. Chau, Zhuo Feng, Ahmed Awadallah, Jennifer Neville, Nikhil Rao
We show that users spend a lot of ``effort'' on these questions in terms of signals like clicks and session length, and that they are also challenging for GPT-4.
no code implementations • 16 Feb 2024 • Arindam Mitra, Hamed Khanpour, Corby Rosset, Ahmed Awadallah
Ensembling provides a substantial boost in accuracy but at a significant cost increase with multiple calls to the model (e. g., Phi-GSM uses top-48 to boost the performance from 68. 2 to 81. 5).
Ranked #35 on Arithmetic Reasoning on GSM8K
no code implementations • 2 Dec 2023 • Corby Rosset, Guoqing Zheng, Victor Dibia, Ahmed Awadallah, Paul Bennett
The remarkable abilities of large language models (LLMs) like GPT-4 partially stem from post-training processes like Reinforcement Learning from Human Feedback (RLHF) involving human preferences encoded in a reward model.
no code implementations • 18 Nov 2023 • Arindam Mitra, Luciano del Corro, Shweti Mahajan, Andres Codas, Clarisse Simoes, Sahaj Agarwal, Xuxi Chen, Anastasia Razdaibiedina, Erik Jones, Kriti Aggarwal, Hamid Palangi, Guoqing Zheng, Corby Rosset, Hamed Khanpour, Ahmed Awadallah
Research on training small LMs has often relied on imitation learning to replicate the output of more capable models.
Ranked #1 on Crass AI on BIG-bench
no code implementations • 14 Nov 2023 • Daniel Campos, Surya Kallumadi, Corby Rosset, Cheng Xiang Zhai, Alessandro Magnani
The focus this year was the creation of a reusable collection and evaluation of the impact of the use of metadata and multi-modal data on retrieval accuracy.
1 code implementation • 3 Oct 2023 • Canwen Xu, Corby Rosset, Ethan C. Chau, Luciano del Corro, Shweti Mahajan, Julian McAuley, Jennifer Neville, Ahmed Hassan Awadallah, Nikhil Rao
Remarkably, our automatic contrastive post-training further improves the performance of Orca, already a state-of-the-art instruction learning model tuned with GPT-4 outputs, to outperform ChatGPT.
no code implementations • 3 Oct 2023 • Guanghui Qin, Corby Rosset, Ethan C. Chau, Nikhil Rao, Benjamin Van Durme
Standard Transformer-based language models (LMs) scale poorly to long contexts.
no code implementations • 7 Feb 2023 • Suyu Ge, Chenyan Xiong, Corby Rosset, Arnold Overwijk, Jiawei Han, Paul Bennett
In this paper we improve the zero-shot generalization ability of language models via Mixture-Of-Memory Augmentation (MoMA), a mechanism that retrieves augmentation documents from multiple information corpora ("external memories"), with the option to "plug in" new memory at inference time.
no code implementations • 30 Jan 2023 • Zhenduo Wang, Yuancheng Tu, Corby Rosset, Nick Craswell, Ming Wu, Qingyao Ai
In this work, we innovatively explore generating clarifying questions in a zero-shot setting to overcome the cold start problem and we propose a constrained clarifying question generation system which uses both question templates and query facets to guide the effective and precise question generation.
no code implementations • 29 Jun 2020 • Corby Rosset, Chenyan Xiong, Minh Phan, Xia Song, Paul Bennett, Saurabh Tiwary
How much knowledge do pretrained language models hold?
1 code implementation • ICLR 2020 • Chen Zhao, Chenyan Xiong, Corby Rosset, Xia Song, Paul Bennett, Saurabh Tiwary
Transformers have achieved new heights modeling natural language as a sequence of text tokens.
Ranked #42 on Question Answering on HotpotQA
no code implementations • 24 Jul 2019 • Hongfei Zhang, Xia Song, Chenyan Xiong, Corby Rosset, Paul N. Bennett, Nick Craswell, Saurabh Tiwary
This paper presents GEneric iNtent Encoder (GEN Encoder) which learns a distributed representation space for user intent in search.
no code implementations • 8 Jul 2019 • Bhaskar Mitra, Corby Rosset, David Hawking, Nick Craswell, Fernando Diaz, Emine Yilmaz
Deep neural IR models, in contrast, compare the whole query to the document and are, therefore, typically employed only for late stage re-ranking.
no code implementations • 15 Apr 2019 • Corby Rosset, Bhaskar Mitra, Chenyan Xiong, Nick Craswell, Xia Song, Saurabh Tiwary
The training of these models involve a search for appropriate parameter values based on large quantities of labeled examples.
no code implementations • 12 Apr 2018 • Corby Rosset, Damien Jose, Gargi Ghosh, Bhaskar Mitra, Saurabh Tiwary
In web search, typically a candidate generation step selects a small set of documents---from collections containing as many as billions of web pages---that are subsequently ranked and pruned before being presented to the user.