The Erosion of LLM Signatures: Can We Still Distinguish Human and LLM-Generated Scientific Ideas After Iterative Paraphrasing?

With the increasing reliance on LLMs as research agents, distinguishing between LLM and human-generated ideas has become crucial for understanding the cognitive nuances of LLMs' research capabilities. While detecting LLM-generated text has been extensively studied, distinguishing human vs LLM-generated scientific idea remains an unexplored area. In this work, we systematically evaluate the ability of state-of-the-art (SOTA) machine learning models to differentiate between human and LLM-generated ideas, particularly after successive paraphrasing stages. Our findings highlight the challenges SOTA models face in source attribution, with detection performance declining by an average of 25.4\% after five consecutive paraphrasing stages. Additionally, we demonstrate that incorporating the research problem as contextual information improves detection performance by up to 2.97%. Notably, our analysis reveals that detection algorithms struggle significantly when ideas are paraphrased into a simplified, non-expert style, contributing the most to the erosion of distinguishable LLM signatures.

Key Contributions

• First systematic evaluation of SOTA detectors distinguishing human vs LLM-generated scientific ideas across successive paraphrasing stages
• Demonstrates that iterative paraphrasing causes a 25.4% average detection performance drop, with simplified non-expert style contributing most to LLM signature erosion
• Shows that incorporating research problem context improves detection by up to 2.97%, suggesting semantic grounding helps attribution

🛡️ Threat Analysis

Details

Similar Papers