Ehud Reiter: four types of LLM-written papers, and why checking matters
EhudReiter · x · 2026-11-03
NLG researcher Ehud Reiter publishes a blog arguing that debates about "LLM-written papers" should be broken into concrete types. Type A: LLM used for brainstorming, literature search, experimental-design advice and writing polish, with authors verifying every output (e.g., actually reading suggested references) — widely seen as acceptable, including under ACL guidance. Type B: same uses but without author verification, e.g., pasting LLM summaries of papers into reviews without reading them. He argues scenario-specific discussion beats generic arguments.
More from Safety
- Red teamer: agentic attacks don't need zero-days, they exploit your security debt — AccBalanced · 2026-11-03
- Questions mount as Anthropic reportedly withheld CVE-2026-98432 exploit details from Ubuntu — jedisct1 · 2026-11-03
- UK AISI spent just £3.34M on compute in 2025-26; researcher says it needs 1000x — HaydnBelfield · 2026-11-03
- Reward hacking risks when agents optimize inference, new article warns — AccBalanced · 2026-11-03
- Railway engineer admits his AI agent probed a competitor's infrastructure, then pulled the plug — AccBalanced · 2026-11-03
- Hacking Prusa printers: 14 findings, ~23 defects, and a full RCE chain — jedisct1 · 2026-11-03