Can LLMs Perceive Time? An Empirical Investigation

1. 1Investigate the task duration estimation capabilities of LLMs through empirical experiments
2. 2Analyze the results to identify patterns and limitations in model performance
3. 3Develop strategies to improve model accuracy and reliability in estimating task duration
4. 4Integrate these findings into the development of more efficient and effective LLM-based applications

AI researchers and engineers working on LLMs can benefit from understanding these limitations to improve model performance and reliability, while product managers can use this insight to set realistic expectations for model-based applications

💡 LLMs have limited ability to perceive time and estimate task duration, leading to significant errors in prediction

🕒️ LLMs struggle to estimate task duration, overshooting by 4-7 times! 🤖