AI Sandbagging: Language Models can Strategically Underperform on Evaluations

The problem with metrics is a fundamental problem for AI

Original

Thomas, R. and Uminsky, D. (2020) · 2002

Earlier work this paper cites.

Measuring massive multitask language understanding

Original

Hendrycks, D., Burns, C., Basart, S., Zou, A., Mazeika, M., Song, D., and Steinhardt, J. (2020) · 2009

Earlier work this paper cites.

Towards formal definitions of blameworthiness, intention, and moral responsibility

Halpern, J. and Kleiman-Weiner, M. (2018) · 2018

Earlier work this paper cites.

Commonsenseqa: A question answering challenge targeting commonsense knowledge

Original

Talmor, A., Herzig, J., Lourie, N., and Berant, J. (2018) · 2018

Earlier work this paper cites.

Language Models are Few-Shot Learners

Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D. M., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A., Sutskever, I., and Amodei, D. (2020) · 2020

Earlier work this paper cites.

LoRA: Low-Rank Adaptation of Large Language Models

Hu, E. J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W. (2021) · 2021

Earlier work this paper cites.

Risks from Learned Optimization in Advanced Machine Learning Systems

Hubinger, E., van Merwijk, C., Mikulik, V., Skalse, J., and Garrabrant, S. (2021) · 2021

Earlier work this paper cites.

Training compute-optimal large language models

Original

Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., Casas, D. d. L., Hendricks, L. A., Welbl, J., Clark, A., et al. (2022) · 2022

Earlier work this paper cites.

The alignment problem from a deep learning perspective

Original

Ngo, R., Chan, L., and Mindermann, S. (2022) · 2022

Earlier work this paper cites.

Discovering Language Model Behaviors with Model-Written Evaluations

Perez, E., Ringer, S., Lukošiūtė, K., Nguyen, K., Chen, E., Heiner, S., Pettit, C., Olsson, C., Kundu, S., Kadavath, S., Jones, A., Chen, A., Mann, B., Israel, B., Seethor, B., McKinnon, C., Olah, C., Yan, D., Amodei, D., Amodei, D., Drain, D., Li, D., Tran-Johnson, E., Khundadze, G., Kernion, J., Landis, J., Kerr, J., Mueller, J., Hyun, J., Landau, J., Ndousse, K., Goldberg, L., Lovitt, L., Lucas, M., Sellitto, M., Zhang, M., Kingsland, N., Elhage, N., Joseph, N., Mercado, N., DasSarma, N., Rausch, O., Larson, R., McCandlish, S., Johnston, S., Kravec, S., Showk, S. E., Lanham, T., Telleen-Lawton, T., Brown, T., Henighan, T., Hume, T., Bai, Y., Hatfield-Dodds, Z., Clark, J., Bowman, S. R., Askell, A., Grosse, R., Hernandez, D., Ganguli, D., Hubinger, E., Schiefer, N., and Kaplan, J. (2022) · 2022

Earlier work this paper cites.

Anthropic’s Responsible Scaling Policy

Anthropic (2023) · 2023

Earlier work this paper cites.

Definitions of intent suitable for algorithms

Ashton, H. (2023) · 2023

Earlier work this paper cites.

Eight Things to Know about Large Language Models

Bowman, S. R. (2023) · 2023

Earlier work this paper cites.

Poisoning Web-Scale Training Datasets is Practical

Carlini, N., Jagielski, M., Choquette-Choo, C. A., Paleka, D., Pearce, W., Anderson, H., Terzis, A., Thomas, K., and Tramèr, F. (2023) · 2023

Earlier work this paper cites.

Scheming AIs: Will AIs fake alignment during training in order to get power?

Original

Carlsmith, J. (2023) · 2023

Earlier work this paper cites.

AI control: Improving safety despite intentional subversion

Original

Greenblatt, R., Shlegeris, B., Sachan, K., and Roger, F. (2023) · 2023

Earlier work this paper cites.

Language models represent space and time

Original

Gurnee, W. and Tegmark, M. (2023) · 2023

Earlier work this paper cites.

Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks

Original

Jain, S., Kirk, R., Lubana, E. S., Dick, R. P., Tanaka, H., Grefenstette, E., Rockt"aschel, T., and Krueger, D. S. (2023) · 2023

Earlier work this paper cites.

Mistral 7B

Jiang, A. Q., Sablayrolles, A., Mensch, A., Bamford, C., Chaplot, D. S., de las Casas, D., Bressand, F., Lengyel, G., Lample, G., Saulnier, L., Lavaud, L. R., Lachaux, M.-A., Stock, P., Scao, T. L., Lavril, T., Wang, T., Lacroix, T., and Sayed, W. E. (2023) · 2023

Earlier work this paper cites.

Governing General Purpose AI

Maham, P. and Küspert, S. (2023) · 2023

Earlier work this paper cites.

Preparedness Framework (Beta)

OpenAI (2023) · 2023

Earlier work this paper cites.

How to catch an AI liar: Lie detection in black-box LLMs by asking unrelated questions

Original

Pacchiardi, L., Chan, A. J., Mindermann, S., Moscovitz, I., Pan, A. Y., Gal, Y., Evans, O., and Brauner, J. (2023) · 2023

Earlier work this paper cites.

Pretraining on the test set is all you need

Original

Schaeffer, R. (2023) · 2023

Earlier work this paper cites.

Practices for Governing Agentic AI Systems

Shavit, Y., Agarwal, S., Brundage, M., Adler, S., O’Keefe, C., Campbell, R., Lee, T., Mishkin, P., Eloundou, T., Hickey, A., et al. (2023) · 2023

Earlier work this paper cites.

Model evaluation for extreme risks

Original

Shevlane, T., Farquhar, S., Garfinkel, B., Phuong, M., Whittlestone, J., Leung, J., Kokotajlo, D., Marchal, N., Anderljung, M., Kolt, N., et al. (2023) · 2023

Earlier work this paper cites.

Llama 2: Open Foundation and Fine-Tuned Chat Models

Touvron, H., Martin, L., Stone, K., Albert, P., Almahairi, A., Babaei, Y., Bashlykov, N., Batra, S., Bhargava, P., Bhosale, S., Bikel, D., Blecher, L., Ferrer, C. C., Chen, M., Cucurull, G., Esiobu, D., Fernandes, J., Fu, J., Fu, W., Fuller, B., Gao, C., Goswami, V., Goyal, N., Hartshorn, A., Hosseini, S., Hou, R., Inan, H., Kardas, M., Kerkez, V., Khabsa, M., Kloumann, I., Korenev, A., Koura, P. S., Lachaux, M.-A., Lavril, T., Lee, J., Liskovich, D., Lu, Y., Mao, Y., Martinet, X., Mihaylov, T., Mishra, P., Molybog, I., Nie, Y., Poulton, A., Reizenstein, J., Rungta, R., Saladi, K., Schelten, A., Silva, R., Smith, E. M., Subramanian, R., Tan, X. E., Tang, B., Taylor, R., Williams, A., Kuan, J. X., Xu, P., Yan, Z., Zarov, I., Zhang, Y., Fan, A., Kambadur, M., Narang, S., Rodriguez, A., Stojnic, R., Edunov, S., and Scialom, T. (2023) · 2023

Cited alongside, same era.

Poisoning Language Models During Instruction Tuning

Wan, A., Wallace, E., Shen, S., and Klein, D. (2023) · 2023

Cited alongside, same era.

Voyager: An open-ended embodied agent with large language models

Original

Wang, G., Xie, Y., Jiang, Y., Mandlekar, A., Xiao, C., Zhu, Y., Fan, L., and Anandkumar, A. (2023) · 2023

Cited alongside, same era.

Inspect An open-source framework for large language model evaluations

AI Safety Institute (2024) · 2024

Cited alongside, same era.

The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning

Original

Li, N., Pan, A., Gopal, A., Yue, S., Berrios, D., Gatti, A., Li, J. D., Dombrowski, A.-K., Goel, S., Phan, L., et al. (2024) · 2024

Closest in time.

Eight Methods to Evaluate Robust Unlearning in LLMs

Original

Lynch, A., Guo, P., Ewart, A., Casper, S., and Hadfield-Menell, D. (2024) · 2024

Closest in time.

Eliciting Latent Knowledge from Quirky Language Models

Mallen, A., Brumley, M., Kharchenko, J., and Belrose, N. (2024) · 2024

Closest in time.

Trojan Detection in Large Language Models: Insights from The Trojan Detection Challenge

Original

Maloyan, N., Verma, E., Nutfullin, B., and Ashinov, B. (2024) · 2024

Closest in time.

Introducing Meta Llama 3: The most capable openly available LLM to date

Meta (2024) · 2024