▲The Inference-Cost Efficiency Race

Score history

Daily conviction score, 0 to 100. Higher means the thesis is more strongly corroborated.

Now 100 · +1 since Oct 9 · ranged 99 to 100

Showing the last few days. Unlock full score history.

Why the conviction moved

• Oct 10
  Strengthened +2

  A new arXiv paper lets the acting model decide which earlier tool results to compress, reversibly. That targets tool-output bloat in context, which drives per-query tokens and KV-cache cost in agent workloads.

• Oct 9
  Strengthened +3

  Letting an agent control what it forgets cut a debugging session's prompt tokens from 912,000 to 232,000 without retraining. That directly lowers per-session serving cost for long-running agents.

Showing the last 2 days. Unlock the full record.

Source trail

• Supporting · October 10, 2026

  New arXiv papers target two agent bottlenecks: tool-output bloat in context and lack of enterprise training data

  A new arXiv paper lets the acting model decide which earlier tool results to compress, reversibly. That targets tool-output bloat in context, which drives per-query tokens and KV-cache cost in agent workloads.

  arXiv cs.AI

• Supporting · October 9, 2026

  Agent-Controlled Forgetting Cuts a Debugging Session's Prompt Tokens From 912,000 to 232,000 Without Retraining

  Letting an agent control what it forgets cut a debugging session's prompt tokens from 912,000 to 232,000 without retraining. That directly lowers per-session serving cost for long-running agents.

  arXiv cs.AI