Harnessing Chain-of-Thought Metadata for Task Routing and Adversarial Prompt Detection
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913762647212032 |
|---|---|
| author | Marinelli, Ryan Pichlmeier, Josef Bisztray, Tamas |
| author_facet | Marinelli, Ryan Pichlmeier, Josef Bisztray, Tamas |
| contents | In this work, we propose a metric called Number of Thoughts (NofT) to determine the difficulty of tasks pre-prompting and support Large Language Models (LLMs) in production contexts. By setting thresholds based on the number of thoughts, this metric can discern the difficulty of prompts and support more effective prompt routing. A 2% decrease in latency is achieved when routing prompts from the MathInstruct dataset through quantized, distilled versions of Deepseek with 1.7 billion, 7 billion, and 14 billion parameters. Moreover, this metric can be used to detect adversarial prompts used in prompt injection attacks with high efficacy. The Number of Thoughts can inform a classifier that achieves 95% accuracy in adversarial prompt detection. Our experiments ad datasets used are available on our GitHub page: https://github.com/rymarinelli/Number_Of_Thoughts/tree/main. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_21464 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Harnessing Chain-of-Thought Metadata for Task Routing and Adversarial Prompt Detection Marinelli, Ryan Pichlmeier, Josef Bisztray, Tamas Computation and Language Artificial Intelligence Performance In this work, we propose a metric called Number of Thoughts (NofT) to determine the difficulty of tasks pre-prompting and support Large Language Models (LLMs) in production contexts. By setting thresholds based on the number of thoughts, this metric can discern the difficulty of prompts and support more effective prompt routing. A 2% decrease in latency is achieved when routing prompts from the MathInstruct dataset through quantized, distilled versions of Deepseek with 1.7 billion, 7 billion, and 14 billion parameters. Moreover, this metric can be used to detect adversarial prompts used in prompt injection attacks with high efficacy. The Number of Thoughts can inform a classifier that achieves 95% accuracy in adversarial prompt detection. Our experiments ad datasets used are available on our GitHub page: https://github.com/rymarinelli/Number_Of_Thoughts/tree/main. |
| title | Harnessing Chain-of-Thought Metadata for Task Routing and Adversarial Prompt Detection |
| topic | Computation and Language Artificial Intelligence Performance |
| url | https://arxiv.org/abs/2503.21464 |