Fetching the paper…

Reading the bibliography…

2024

Defending Against Indirect Prompt Injection Attacks With Spotlighting

Hines, Keegan, Lopez, Gary, Hall, Matthew et al.

Original
PDF
Post on X

Understand

Large Language Models (LLMs), while powerful, are built and trained to process a single text input.

In common applications, multiple inputs can be processed by concatenating them together into a single stream of text.
However, the LLM is unable to distinguish which sections of prompt belong to various input sources.
Indirect prompt injection attacks take advantage of this vulnerability by embedding adversarial instructions into untrusted data being processed alongside user commands.

Defending Against Indirect Prompt Injection Attacks With Spotlighting · Around

Built on

A. L. Maas, R. E. Daly, P. T. Pham, D. Huang, A. Y. Ng, C. Potts, “Learning Word Vectors for Sentiment Analysis,” in
2011
Earlier work this paper cites.
P. Rajpurkar, J. Zhang, K. Lopyrev, P. Liang, “SQuAD: 100,000+ Questions for Machine Comprehension of Text,”
Original
2016
Earlier work this paper cites.
A. Wang, Y. Pruksachatkun, N. Nangia, A. Singh, J. Michael, F. Hill, O. Levy, S. R. Bowman, “SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems,”
Original
2020
Earlier work this paper cites.
T. B. Brown, B. Mann, N. Ryder, M. Subbiah, J. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell,
Original
2020
Earlier work this paper cites.
Y. Bai, S. Kadavath, S. Kundu, A. Askell, J. Kernion, A. Jones, A. Chen, A. Goldie, A. Mirhoseini, C. McKinnon,
Original
2022
Earlier work this paper cites.
A. Chowdhery, S. Narang, J. Devlin, M. Bosma, G. Mishra, A. Roberts, P. Barham, H. W. Chung, C. Sutton, S. Gehrmann,
Original
2022
Earlier work this paper cites.
L. Ouyang, S. Toyer, C. Donahue, J. Rahim, Y. Bao, J. Wu, H. He, Z. Tung, A. Chaganty, P. Liang, C. D. Manning, J. Pennington, A. Radford, D. Amodei,
Original
2022
Earlier work this paper cites.

Similar

B. Roziere, J. Gehring, F. Gloeckle, S. Sootla, I. Gat, X. E. Tan, Y. Adi, J. Liu, T. Remez, J. Rapin,
Original
2023
Cited alongside, same era.
J. Yi, Y. Xie, B. Zhu, K. Hines, E. Kiciman, G. Sun, X. Xie, F. Wu, “Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models”,
Original
2023
Cited alongside, same era.
OpenAI, “GPT-4 Technical Report,”
Original
2023
Cited alongside, same era.
H. Touvron, L. Martin, K. Stone, P. Albert, A. Almahairi, Y. Babaei, N. Bashlykov, S. Batra, P. Bhargava, S. Bhosale,
Original
2023
Cited alongside, same era.
K. Greshake, S. Abdelnabi, S. Mishra, C. Endres, T. Holz, M. Fritz, “More than you’ve asked for: A Comprehensive Analysis of Novel Prompt Injection Threats to Application-Integrated Large Language Models,”

Then

A. Zou, Z. Wang, N. Carlini, M. Nasr, J. Z. Kolter, M. Fredrikson,
Original
2023
Later among the works it cites.
International Telecommunication Union, “Q Series: Switching and Signalling No. 5,” 1988. [Online]. Available: https://www.itu.int/rec/T-REC-Q.140-Q.180-198811-I/en. [Accessed: Feb. 2, 2024]
2024
Closest in time.
International Telecommunication Union, “Q Series: Switching and Signalling No. 5,” 1988. [Online]. Available: https://www.itu.int/rec/T-REC-Q.140-Q.180-198811-I/en. [Accessed: Feb. 2, 2024]
2024
Closest in time.
International Telecommunication Union, “Q Series: Switching and Signalling No. 6,” 1988. [Online]. Available: https://www.itu.int/rec/T-REC-Q.251-Q.300-198811-I/en. [Accessed: Feb. 2, 2024]
2024
Closest in time.
K. Greshake, “How We Broke LLMs: Indirect Prompt Injection,”

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Original

2023

Cited alongside, same era.

J. Wei, X. Wang, D. Schuurmans, M. Bosma, B. Ichter, F. Xia, E. Chi, Q. Le, D. Zhou,

2023

Cited alongside, same era.

S. Yao, D. Yu, J. Zhao, I. Shafran, T. L. Griffiths, Y. Cao, K. Narasimhan,

2023

Cited alongside, same era.

2024

Closest in time.

Wunderwuzzi, “Hacking Google Bard - From Prompt Injection to Data Exfiltration,”

2024

Closest in time.

Anthropic Team, “Core Views on AI Safety: When, Why, What, and How,” 2023. [Online]. Available: https://www.anthropic.com/news/core-views-on-ai-safety. [Accessed: Feb. 21, 2024]

2024

Closest in time.