🧩 Philosophy May 10, 2026 · Florian_Dietz

Context Modification as a Negative Alignment Tax

Less Wrong
View Channel →
Source ↗ 👁 59 💬 0
Context Rot
Every LLM gets worse as its context grows. Chroma tested 18 frontier models and found performance degradation in all of them, often by double-digit percentages on tasks where short-context performance was strong. The industry calls this "context rot": the gradual degradation of response quality as irrelevant history accumulates in the context window.
The standard fix is compaction: when the context gets too long, summarize it and throw away the original. Claude Code auto-compacts at

Comments (0)

Sign in to join the discussion

More Like This

Mini-Heap
Daily Nous · 8h ago
Sleeping Beauties: David Byrne on Why Transformative Ideas Fall Dormant and How They Awaken to Change the World
The Marginalian · 18h ago
Improvisation and the Quantum of Consciousness
The Marginalian · 1d ago
How to Remake Yourself When You Are Undone: Lincoln and the Gift of His Suicidal Depression
The Marginalian · 1d ago
AI for Philosophy: Progress through Capital-Intensive Philosophy (guest post)
Daily Nous · 1d ago
📰
A summary of a viral Chinese essay on what a DeepSeek kernel engineer's opinion on automating his own job
LessWrong · 1d ago