ai-research-survey

Systematic scan of agentic development research. What's signal, what's noise.
git clone https://git.shiptheloop.com/ai-research-survey.git
Log | Files | Refs

paper_type.json (381B)


      1 {
      2   "paper_type": "empirical",
      3   "reason": "The paper runs systematic experiments on 10 LLMs using semantic-preserving mutations to empirically demonstrate that benchmark rankings are unstable and inflated, with performance drops up to 40-50%—the primary contribution is these experimental findings about benchmark vulnerability, not the dynamic benchmarking framework itself."
      4 }

Impressum · Datenschutz