Arguably those sections are exactly where LLM use would be least problematic; i.e., summaries, future work, etc. Those sections are also where language proficiency tends to have the most outsized effect, where LLMs might be used to achieve a degree of clarity and brevity that’s otherwise out of reach. The fact that there was less evidence of use in methods and results would be a good sign WRT usage trends, I would think.
Arguably those sections are exactly where LLM use would be least problematic; i.e., summaries, future work, etc. Those sections are also where language proficiency tends to have the most outsized effect, where LLMs might be used to achieve a degree of clarity and brevity that’s otherwise out of reach. The fact that there was less evidence of use in methods and results would be a good sign WRT usage trends, I would think.