Analyzing the LLM Slop Trend in Submissions
NeelNanda5 · x · 2026-07-15
The author mentions having organized 3 mech interp workshops, inadvertently accumulating a dataset of submissions perfect for analyzing the upward trend of "LLM slop".
@andyarditi's analysis focuses on:
- The actual volume of AI slop accepted
- How this situation has evolved since 2024
- Whether the review process is genuinely "all noise"
The author emphasizes that this is a post-hoc analysis and that Pangram was not used when making their decisions.
Related event: Data Confirms Worsening LLM Slop in Academic Submissions(2 posts)→
More from Research
- Open-source decision models sold their own base in Red Alert 2 — then learned until no human could beat them — heathercmiller · 2026-10-27
- DeepMind paper: LLMs can't sample randomly on their own — pass random numbers from code instead — rohanpaul_ai · 2026-10-27
- World Model Assigns Weight and Friction to Objects, Then Simulates with a Physics Engine — jonstephens85 · 2026-10-27
- Token Collapse: Study Finds Graph Soft Prompting Tokens in LLMs Fail to Encode Structure — Sapienza · 2026-10-27
- Mathathon 2.0 launches with Terence Tao's SAIR, $1.5M in compute for teams — _sathvikr · 2026-10-27
- The 'sweet lesson' is real: two kinds of ML problems need opposite strategies — mike64_t · 2026-10-27