(long-context agent monitoring benchmark — not explicitly named in abstract) D

AI benchmark · Long Context

Publisher
Paper Arxiv
Domains
Long Context
Tier
D why
Snapshot

What is (long-context agent monitoring benchmark — not explicitly named in abstract)?

Monitoring coding agents for dangerous behavior using language models requires classifying transcripts that often exceed 500K tokens, but prior agent monitoring benchmarks rarely contain transcripts longer than 100K tokens. We show that when used as classifiers, current frontier models fail to notice dangerous actions more often in longer transcripts. In particular, on a dataset that requires identifying when a coding agent takes a subtly dangerous action, Opus 4.6, GPT 5.4, and Gemini 3.1 miss these actions $2\times$ to $30\times$ more often when they occur after 800K tokens of benign activity than when they occur on their own. We also show that these weaknesses can be partially mitigated with prompting techniques such as periodic reminders throughout the transcript and may be mitigated further with better post-training. Monitor evaluations that do not consider long-context degradation are likely overestimating monitor performance.

Tier D

D

Long tail: little or no usage data yet.

How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.

  • 3 models scored
  • no proper name (a paper title or description)
  • in curated RL directories (1 pt)
How we got to tier D
  1. Adoption: 3 models scored; no proper name (a paper title or description). Tier D.
  2. Not yet read on X.
  3. In curated RL directories: 1 point, a floor of none.
  4. Not yet confirmed by a line from its own site or paper.

A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.

Not on the map yet: no line from its own site or paper places it in a domain.

More Long Context AI benchmarks

Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-11 snapshot.