Fetching the paper…
Reading the bibliography…
In the absence of abundant reliable annotations for challenging tasks and contexts, how can we expand the frontier of LLM capabilities with potentially wrong answers? We focus on two research questions: (1) Can LLMs generate reliable preferences among wrong options? And if so, (2) Would alignment with such wrong-over-wrong preferences be helpful? We employ methods based on self-consistency, token probabilities, and LLM-as-a-judge to elicit wrong-over-wrong preferences, and fine-tune language models with preference optimization approaches using these synthesized preferences.
Pearson’s correlation coefficient
Philip Sedgwick · 2012
Earlier work this paper cites.
On calibration of modern neural networks
Chuan Guo, Geoff Pleiss, Yu Sun, and Kilian Q Weinberger · 2017
Earlier work this paper cites.
Proximal policy optimization algorithms
John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov · 2017
Earlier work this paper cites.
Bohb: Robust and efficient hyperparameter optimization at scale
Stefan Falkner, Aaron Klein, and Frank Hutter · 2018
Earlier work this paper cites.
Hellaswag: Can a machine really finish your sentence?
Rowan Zellers, Ari Holtzman, Yonatan Bisk, Ali Farhadi, and Yejin Choi · 2019
Earlier work this paper cites.
Fine-tuning language models from human preferences
Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving · 2019
Earlier work this paper cites.
Learning to summarize from human feedback
Nisan Stiennon, Long Ouyang, Jeff Wu, Daniel M Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul Christiano · 2020
Earlier work this paper cites.
Transformers: State-of-the-art natural language processing
Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Rémi Louf, Morgan Funtowicz, et al · 2020
Earlier work this paper cites.
On the dangers of stochastic parrots: Can language models be too big?
Emily M Bender, Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell · 2021
Earlier work this paper cites.
Stockfish, 2021
T. Romstad, M. Costalba, J. Kiiski, G. Linscott, S. Nicolet, S. Geschwentner, and J. VandeVondele · 2021
Earlier work this paper cites.
Naturalproofs: Mathematical theorem proving in natural language
Sean Welleck, Jiacheng Liu, Ronan Le Bras, Hannaneh Hajishirzi, Yejin Choi, and Kyunghyun Cho · 2021
Earlier work this paper cites.
Detoxifying language models risks marginalizing minority voices
Albert Xu, Eshaan Pathak, Eric Wallace, Suchin Gururangan, Maarten Sap, and Dan Klein · 2021
Earlier work this paper cites.
Constitutional ai: Harmlessness from ai feedback
Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, et al · 2022
Earlier work this paper cites.
Measuring progress on scalable oversight for large language models
Samuel R Bowman, Jeeyoon Hyun, Ethan Perez, Edwin Chen, Craig Pettit, Scott Heiner, Kamilė Lukošiūtė, Amanda Askell, Andy Jones, Anna Chen, et al · 2022
Earlier work this paper cites.
Teaching models to express their uncertainty in words
Stephanie Lin, Jacob Hilton, and Owain Evans · 2022
Earlier work this paper cites.
Training language models to follow instructions with human feedback
Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al · 2022
Earlier work this paper cites.
Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering
Ankit Pal, Logesh Kumar Umapathi, and Malaikannan Sankarasubbu · 2022
Earlier work this paper cites.
Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al · 2023
Earlier work this paper cites.
Benchmarking foundation models with language-model-as-an-examiner
Yushi Bai, Jiahao Ying, Yixin Cao, Xin Lv, Yuze He, Xiaozhi Wang, Jifan Yu, Kaisheng Zeng, Yijia Xiao, Haozhe Lyu, et al · 2023
Earlier work this paper cites.
Code alpaca: An instruction-following llama model for code generation
Sahil Chaudhary · 2023
Earlier work this paper cites.
Selectively answering ambiguous questions
Jeremy Cole, Michael Zhang, Dan Gillick, Julian Eisenschlos, Bhuwan Dhingra, and Jacob Eisenstein · 2023
Earlier work this paper cites.
Qlora: efficient finetuning of quantized llms
Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer · 2023
Earlier work this paper cites.
Knowledge crosswords: Geometric reasoning over structured knowledge with large language models
Wenxuan Ding, Shangbin Feng, Yuhan Liu, Zhaoxuan Tan, Vidhisha Balachandran, Tianxing He, and Yulia Tsvetkov · 2023
Earlier work this paper cites.
Alpacafarm: a simulation framework for methods that learn from human feedback
Yann Dubois, Xuechen Li, Rohan Taori, Tianyi Zhang, Ishaan Gulrajani, Jimmy Ba, Carlos Guestrin, Percy Liang, and Tatsunori B Hashimoto · 2023
Earlier work this paper cites.
From pretraining data to language models to downstream tasks: Tracking the trails of political biases leading to unfair nlp models
Shangbin Feng, Chan Young Park, Yuhan Liu, and Yulia Tsvetkov · 2023
Earlier work this paper cites.
Camels in a changing climate: Enhancing lm adaptation with tulu 2
Hamish Ivison, Yizhong Wang, Valentina Pyatkin, Nathan Lambert, Matthew Peters, Pradeep Dasigi, Joel Jang, David Wadden, Noah A Smith, Iz Beltagy, et al · 2023
Earlier work this paper cites.
Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al · 2023
Earlier work this paper cites.
Openassistant conversations-democratizing large language model alignment
Andreas Köpf, Yannic Kilcher, Dimitri von Rütte, Sotiris Anagnostidis, Zhi-Rui Tam, Keith Stevens, Abdullah Barhoum, Nguyen Minh Duc, Oliver Stanley, Richárd Nagyfi, et al · 2023
Cited alongside, same era.
Rlaif: Scaling reinforcement learning from human feedback with ai feedback
Harrison Lee, Samrat Phatale, Hassan Mansoor, Thomas Mesnard, Johan Ferret, Kellie Lu, Colton Bishop, Ethan Hall, Victor Carbune, Abhinav Rastogi, et al · 2023
Cited alongside, same era.
Lichess database
Lichess Team · 2023
Cited alongside, same era.
Vera: A general-purpose plausibility estimation model for commonsense statements
Jiacheng Liu, Wenya Wang, Dianzhuo Wang, Noah A Smith, Yejin Choi, and Hannaneh Hajishirzi · 2023
Cited alongside, same era.
Selfcheckgpt: Zero-resource black-box hallucination detection for generative large language models
Potsawee Manakul, Adian Liusie, and Mark Gales · 2023
Cited alongside, same era.
Safe rlhf: Safe reinforcement learning from human feedback
Josef Dai, Xuehai Pan, Ruiyang Sun, Jiaming Ji, Xinbo Xu, Mickel Liu, Yizhou Wang, and Yaodong Yang · 2024
Closest in time.
Quality diversity through human feedback: Towards open-ended diversity-driven optimization
Li Ding, Jenny Zhang, Jeff Clune, Lee Spector, and Joel Lehman · 2024
Closest in time.
Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al · 2024
Closest in time.
Complex reasoning over logical queries on commonsense knowledge graphs
Tianqing Fang, Zeming Chen, Yangqiu Song, and Antoine Bosselut · 2024
Closest in time.
Modular pluralism: Pluralistic alignment via multi-llm collaboration
Shangbin Feng, Taylor Sorensen, Yuhan Liu, Jillian Fisher, Chan Young Park, Yejin Choi, and Yulia Tsvetkov · 2024
alphaXiv searches the wider corpus for related work and actual follow-ups.
alphaXiv is searching for related work…
Factscore: Fine-grained atomic evaluation of factual precision in long form text generation
Sewon Min, Kalpesh Krishna, Xinxi Lyu, Mike Lewis, Wen-tau Yih, Pang Koh, Mohit Iyyer, Luke Zettlemoyer, and Hannaneh Hajishirzi · 2023
Cited alongside, same era.
Direct preference optimization: your language model is secretly a reward model
Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D Manning, and Chelsea Finn · 2023
Cited alongside, same era.
Gemini: a family of highly capable multimodal models
Gemini Team, Rohan Anil, Sebastian Borgeaud, Yonghui Wu, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M Dai, Anja Hauth, et al · 2023
Cited alongside, same era.
Just ask for calibration: Strategies for eliciting calibrated confidence scores from language models fine-tuned with human feedback
Katherine Tian, Eric Mitchell, Allan Zhou, Archit Sharma, Rafael Rafailov, Huaxiu Yao, Chelsea Finn, and Christopher D Manning · 2023
Cited alongside, same era.
Llama: Open and efficient foundation language models
Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al · 2023
Cited alongside, same era.
Can large language models really improve by self-critiquing their own plans?
Karthik Valmeekam, Matthew Marquez, and Subbarao Kambhampati · 2023
Cited alongside, same era.
Wizardlm: Empowering large language models to follow complex instructions
Can Xu, Qingfeng Sun, Kai Zheng, Xiubo Geng, Pu Zhao, Jiazhan Feng, Chongyang Tao, and Daxin Jiang · 2023
Cited alongside, same era.
Closest in time.
A survey of confidence estimation and calibration in large language models
Jiahui Geng, Fengyu Cai, Yuxia Wang, Heinz Koeppl, Preslav Nakov, and Iryna Gurevych · 2024
Closest in time.
The unreasonable effectiveness of easy training data for hard tasks
Peter Hase, Mohit Bansal, Peter Clark, and Sarah Wiegreffe · 2024
Closest in time.
Orpo: Monolithic preference optimization without reference model
Jiwoo Hong, Noah Lee, and James Thorne · 2024
Closest in time.
Debating with more persuasive llms leads to more truthful answers
Akbir Khan, John Hughes, Dan Valentine, Laura Ruis, Kshitij Sachan, Ansh Radhakrishnan, Edward Grefenstette, Samuel R Bowman, Tim Rocktäschel, and Ethan Perez · 2024
Closest in time.
Learning to trust your feelings: Leveraging self-awareness in llms for hallucination mitigation
Yuxin Liang, Zhuoyang Song, Hao Wang, and Jiaxing Zhang · 2024
Closest in time.
Flame: Factuality-aware alignment for large language models
Sheng-Chieh Lin, Luyu Gao, Barlas Oguz, Wenhan Xiong, Jimmy Lin, Wen-tau Yih, and Xilun Chen · 2024
Closest in time.
Calibrating llm-based evaluator
Yuxuan Liu, Tianchi Yang, Shaohan Huang, Zihan Zhang, Haizhen Huang, Furu Wei, Weiwei Deng, Feng Sun, and Qi Zhang · 2024
Closest in time.
Simpo: Simple preference optimization with a reference-free reward
Yu Meng, Mengzhou Xia, and Danqi Chen · 2024
Closest in time.
Selfcheck: Using LLMs to zero-shot check their own step-by-step reasoning
Ning Miao, Yee Whye Teh, and Tom Rainforth · 2024
Closest in time.
Nash learning from human feedback
Remi Munos, Michal Valko, Daniele Calandriello, Mohammad Gheshlaghi Azar, Mark Rowland, Zhaohan Daniel Guo, Yunhao Tang, Matthieu Geist, Thomas Mesnard, Côme Fiegel, et al · 2024
Closest in time.
Fine-tuning aligned language models compromises safety, even when users do not intend to!
Xiangyu Qi, Yi Zeng, Tinghao Xie, Pin-Yu Chen, Ruoxi Jia, Prateek Mittal, and Peter Henderson · 2024
Closest in time.
A roadmap to pluralistic alignment
Taylor Sorensen, Jared Moore, Jillian Fisher, Mitchell Gordon, Niloofar Mireshghallah, Christopher Michael Rytting, Andre Ye, Liwei Jiang, Ximing Lu, Nouha Dziri, et al · 2024
Closest in time.
Gemma: Open models based on gemini research and technology
Gemma Team, Thomas Mesnard, Cassidy Hardin, Robert Dadashi, Surya Bhupatiraju, Shreya Pathak, Laurent Sifre, Morgane Rivière, Mihir Sanjay Kale, Juliette Love, et al · 2024
Closest in time.
The art of refusal: A survey of abstention in large language models
Bingbing Wen, Jihan Yao, Shangbin Feng, Chenjun Xu, Yulia Tsvetkov, Bill Howe, and Lucy Lu Wang · 2024
Closest in time.
The generative ai paradox:“what it can create, it may not understand”
Peter West, Ximing Lu, Nouha Dziri, Faeze Brahman, Linjie Li, Jena D Hwang, Liwei Jiang, Jillian Fisher, Abhilasha Ravichander, Khyathi Chandu, et al · 2024
Closest in time.
Self-play preference optimization for language model alignment
Yue Wu, Zhiqing Sun, Huizhuo Yuan, Kaixuan Ji, Yiming Yang, and Quanquan Gu · 2024
Closest in time.
Rlcd: Reinforcement learning from contrastive distillation for lm alignment
Kevin Yang, Dan Klein, Asli Celikyilmaz, Nanyun Peng, and Yuandong Tian · 2024
Closest in time.
Self-rewarding language models
Weizhe Yuan, Richard Yuanzhe Pang, Kyunghyun Cho, Xian Li, Sainbayar Sukhbaatar, Jing Xu, and Jason E Weston · 2024
Closest in time.
Can llm graph reasoning generalize beyond pattern memorization?
Yizhuo Zhang, Heng Wang, Shangbin Feng, Zhaoxuan Tan, Xiaochuang Han, Tianxing He, and Yulia Tsvetkov · 2024
Closest in time.
Long is more for alignment: A simple but tough-to-beat baseline for instruction fine-tuning
Hao Zhao, Maksym Andriushchenko, Francesco Croce, and Nicolas Flammarion · 2024
Closest in time.
REFACTOR: Learning to extract theorems from proofs
Jin Peng Zhou, Yuhuai Wu, Qiyang Li, and Roger Baker Grosse · 2024
Closest in time.