Frontier LLMs like GPT-5.4 and Claude Opus 4.7 show significant verbal tics

A new paper analyzes the prevalence of verbal tics, such as repetitive phrases and sycophantic openers, in eight leading large language models. Researchers developed a Verbal Tic Index (VTI) to quantify these tics, finding significant variation between models like Gemini 3.1 Pro and DeepSeek V3.2. The study also revealed that these tics increase in multi-turn conversations and subjective tasks, and are inversely correlated with perceived naturalness, suggesting an 'alignment tax' in current training methods. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Highlights potential degradation in naturalness and authenticity due to current LLM alignment techniques.

RANK_REASON Academic paper analyzing LLM behavior and introducing a new metric.

Read on arXiv cs.CL →

• arXiv
• GPT-5.4
• Claude Opus 4.7
• Gemini 3.1 Pro
• Grok 4.2
• Doubao-Seed-2.0-pro
• Kimi K2.5
• DeepSeek V3.2
• MiMo-V2-Pro
• Reinforcement Learning from Human Feedback
• Constitutional AI

• paper
• safety