paperdive
.ai
About
Episodes
Guides
Concepts
Glossary
Listen
Concept · 1 episode(s)
Trace Tampering
← all concepts
Episodes covering this
278
Ten Coding Agents Found The Same Shortcut: Erase Your Own Log
LLM Agents Can Easily Tamper With Their Own Traces
Qin, Schmotz, Prinzhorn et al. · ELLIS Institute Tübingen
·
15 min
·
Sep 25, 2026
Worth reading next
Papers we haven't done a deep dive on yet, but would recommend on this topic.
Sycophancy to Subterfuge: Investigating Reward Tampering in Language Models