| Recipes for building an open-domain chatbot S Roller arXiv preprint arXiv:2004.13637, 2020 | 1051 | 2020 |
| Blenderbot 3: a deployed conversational agent that continually learns to responsibly engage K Shuster, J Xu, M Komeili, D Ju, EM Smith, S Roller, M Ung, M Chen, ... arXiv preprint arXiv:2208.03188, 2022 | 250 | 2022 |
| Beyond goldfish memory: Long-term open-domain conversation J Xu arXiv preprint arXiv:2107.07567, 2021 | 235 | 2021 |
| Chain-of-verification reduces hallucination in large language models S Dhuliawala, M Komeili, J Xu, R Raileanu, X Li, A Celikyilmaz, J Weston arXiv preprint arXiv:2309.11495, 2023 | 190 | 2023 |
| Recipes for safety in open-domain chatbots J Xu, D Ju, M Li, YL Boureau, J Weston, E Dinan arXiv preprint arXiv:2010.07079, 2020 | 179 | 2020 |
| Self-rewarding language models W Yuan, RY Pang, K Cho, S Sukhbaatar, J Xu, J Weston arXiv preprint arXiv:2401.10020, 2024 | 156 | 2024 |
| Bot-adversarial dialogue for safe conversational agents J Xu, D Ju, M Li, YL Boureau, J Weston, E Dinan Proceedings of the 2021 Conference of the North American Chapter of the …, 2021 | 122 | 2021 |
| Learning new skills after deployment: Improving open-domain internet-driven dialogue with human feedback J Xu, M Ung, M Komeili, K Arora, YL Boureau, J Weston arXiv preprint arXiv:2208.03270, 2022 | 35 | 2022 |
| Some things are more cringe than others: Preference optimization with the pairwise cringe loss J Xu, A Lee, S Sukhbaatar, J Weston arXiv preprint arXiv:2312.16682, 2023 | 33 | 2023 |
| Saferdialogues: Taking feedback gracefully after conversational safety failures M Ung, J Xu, YL Boureau arXiv preprint arXiv:2110.07518, 2021 | 33 | 2021 |
| The cringe loss: Learning what language not to model L Adolphs, T Gao, J Xu, K Shuster, S Sukhbaatar, J Weston arXiv preprint arXiv:2211.05826, 2022 | 26 | 2022 |
| On anytime learning at macroscale L Caccia, J Xu, M Ott, M Ranzato, L Denoyer Conference on Lifelong Learning Agents, 165-182, 2022 | 22 | 2022 |
| When life gives you lemons, make cherryade: Converting feedback from bad responses into good labels W Shi, E Dinan, K Shuster, J Weston, J Xu arXiv preprint arXiv:2210.15893, 2022 | 15 | 2022 |
| Learning from data in the mixed adversarial non-adversarial case: Finding the helpers and ignoring the trolls D Ju, J Xu, YL Boureau, J Weston arXiv preprint arXiv:2208.03295, 2022 | 14 | 2022 |
| Blenderbot 3: a deployed conversational agent that continually learns to responsibly engage, 2022 K Shuster, J Xu, M Komeili, D Ju, EM Smith, S Roller, M Ung, M Chen, ... URL https://arxiv. org/abs/2208.03188, 0 | 14 | |
| Meta-rewarding language models: Self-improving alignment with llm-as-a-meta-judge T Wu, W Yuan, O Golovneva, J Xu, Y Tian, J Jiao, J Weston, S Sukhbaatar arXiv preprint arXiv:2407.19594, 2024 | 9 | 2024 |
| Housing choices, sorting, and the distribution of educational benefits under deferred acceptance J Xu Journal of Public Economic Theory 21 (3), 558-595, 2019 | 9 | 2019 |
| Training models to generate, recognize, and reframe unhelpful thoughts M Maddela, M Ung, J Xu, A Madotto, H Foran, YL Boureau arXiv preprint arXiv:2307.02768, 2023 | 7 | 2023 |
| Improving open language models by learning from organic interactions J Xu, D Ju, J Lane, M Komeili, EM Smith, M Ung, M Behrooz, W Ngan, ... arXiv preprint arXiv:2306.04707, 2023 | 5 | 2023 |
| Distilling system 2 into system 1 P Yu, J Xu, J Weston, I Kulikov arXiv preprint arXiv:2407.06023, 2024 | 4 | 2024 |