API continuous dialog always answers previous questions

I achieve continuous dialogs by storing Chatgpt’s previous responses, but there is often a situation where the new response returns the answer to the previous question. Here is an example of my message:
[
{
“role”: “system”,
“content”: “As an expert annotator with a focus on scientific article content analysis, your role involves ascertaining whether the article meets the set criteria.”
},
{
“role”: “user”,
“content”: “ARTICLE: Ganglion cell layer analysis with deep learning in glaucoma diagnosis. OBJECTIVE: To determine and compare the diagnostic precision in glaucoma of two deep learning models using infrared images of the optic nerve, eye fundus, and the ganglion cell layer (GCL).METHODS: We have selected a sample of normal and glaucoma patients. Three infrared images were registered with a spectral-domain optical coherence tomography (SD-OCT). The first corresponds to the confocal scan image of the fundus, the second is a cut-out of the first centered on the optic nerve, and the third was the SD-OCT image of the GCL. Our deep learning models are developed on the MatLab platform with the ResNet50 and VGG19 pre-trained neural networks.RESULTS: 498 eyes of 298 patients were collected. Of the 498 eyes, 312 are glaucoma and 186 are normal. In the test, the precision of the models was 96% (ResNet50) and 96% (VGG19) for the GCL images, 90% (ResNet50) and 90% (VGG19) for the optic nerve images and 82% (ResNet50) and 84% (VGG19) for the fundus images. The ROC area in the test was 0.96 (ResNet50) and 0.97 (VGG19) for the GCL images, 0.87 (ResNet50) and 0.88 (VGG19) for the optic nerve images, and 0.79 (ResNet50) and 0.81 (VGG19) for the fundus images.CONCLUSIONS: Both deep learning models, applied to the GCL images, achieve high diagnostic precision, sensitivity and specificity in the diagnosis of glaucoma.\n\nPlease ascertain whether the article meets the following criteria: Clinical_Study–A clinical study that involves human participants;\n\nProvide your response in JSON format, as follows:{“whether_meet_Clinical_Study_criteria”: “Yes/No”}\n\nOptions:\n- Yes\n- No\n\nANSWER:”
},
{
“role”: “assistant”,
“content”: {
“whether_meet_Clinical_Study_criteria”: “Yes”
}
},
{
“role”: “user”,
“content”: “Please ascertain whether the article meets the following criteria: Glaucoma–A research study about glaucoma;\n\nProvide your response in JSON format, as follows:{“whether_meet_Glaucoma_criteria”: “Yes/No”}\n\nOptions:\n- Yes\n- No\n\nANSWER:”
},
{
“role”: “assistant”,
“content”: {
“whether_meet_Glaucoma_criteria”: “Yes”
}
},
{
“role”: “user”,
“content”: “Please ascertain whether the article meets the following criteria: Diagnosis–A research study that evaluates methods of detecting disease;\n\nProvide your response in JSON format, as follows:{“whether_meet_Diagnosis_criteria”: “Yes/No”}\n\nOptions:\n- Yes\n- No\n\nANSWER:”
},
{
“role”: “assistant”,
“content”: {
“whether_meet_Clinical_Study_criteria”: “Yes”
}
},
{
“role”: “user”,
“content”: “Please ascertain whether the article meets the following criteria: Deep_Learning_Model–A research study that use models based on Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTMs), Recurrent Neural Networks (RNNs), and Generative Adversarial Networks (GANs) etc;\n\nProvide your response in JSON format, as follows:{“whether_meet_Deep_Learning_Model_criteria”: “Yes/No”}\n\nOptions:\n- Yes\n- No\n\nANSWER:”
}
]

May I ask how I can avoid this situation? I did not find a solution for this problem in the forum.

Doesn’t anyone else have the same problem? It’s too terrible.

I tried to re-organize your prompts:

The first answer bugged me so I asked follow up questions.

2 Likes

Thank you very much for your dedication and answers! I have tried to optimize prompt to solve this problem before and it didn’t work well. But this prompt of yours seems to be even better!

1 Like