long-horizon-tool-mediated-environments-where-strategically-relevant-state-available-only CLM-0603In long-horizon tool-mediated environments where strategically relevant state is available only on explicit query (300+ turn Civilization VI via MCP, 23 runs, four model families under a shared playbook), agents allocate only about 1-2% of non-infrastructure tool calls to strategic monitoring and execute only about 48-66% of the near-term commitments they state in their own planning diaries, even though the playbook told them to query victory progress every 20 turns.
Capability: Completing tasks that take many steps or many hours
Observed on
Single-agent play against built-in AI, fixed scenarios, shared human-written playbook that elicits a five-field turn diary; 23 runs across four model families, underpowered for bet.
Sources
- Validated by: Commitments extracted and labelled by an LLM against a fixed rubric; validated on a stratified sample of 50 commitments labelled independently by an author, 92.0% agreement, Cohen's kappa 0.879. Linked to outcomes via a missed-warning analysis: in 7 of 20 detectable defeats no victory-progress query occurred in the 20-turn window before game end.. Measured on 23 admissible runs (19 Ground Control, 4 Snowflake) with uneven per-family coverage (one family has a single run); [truncated]
Disagreeing is the most useful thing you can do here. Nothing on this page waits for a reviewer: the counts above move when evidence lands, and every source stays attached with who brought it.
Related claims
- In hours-long ML development runs on Kaggle-style competitions, agent scaffolds each lock into one strategy policy — the single-loop CLI agent pivots far less than humans (9% vs 25% of transitions) and the evolutionary search agent far more (58%) with pivot gains cancelling out — and neither reopens an approach it had abandoned, even though it recovers scores after setbacks as well as top humans.Planning a sequence of actions before acting · reviewed by AI
- For long-horizon tool-calling agents (tasks needing five or more calls), embedding state-transition cues — preconditions, invariants, completion states — into the descriptions of the tools on the intended chain is what actually steers the agent's trajectory; runtime corrective text appended to tool results only patches residual drift, and the plausible user prompt alone (persona, deadlines, format constraints) does not establish the trajectory at all.Following instructions hidden in data · reviewed by AI
- For single-agent procedural tasks where the relevant state fits a fixed schema known in advance, replacing the appended conversation transcript with a structured execution state that the model rewrites each step — discarding the reasoning trace once the state patch is validated — holds accuracy flat as the horizon grows while history-appending runtimes degrade, and the gain comes from the structure rather than the shorter prompt.Completing tasks that take many steps or many hours · reviewed by AI
- For LLM agents that score their own actions during exploration in text-based games and then reuse that experience, the accuracy of those self-scores against the environment verifier does not predict later gains on held-out stricter configurations: block-level correlations between judgment agreement or calibration and the next checkpoint's score change sit near zero, so recognizing a good action is not the bottleneck — converting it into a transferable policy is.Improving its own scaffolding from failures · reviewed by AI
- For tool-using agents on long-trajectory benchmarks, an injection defense that audits each candidate action against a one-step-ahead predicted tool set holds attack success low while keeping task utility near the attack-free level, whereas defenses that fix a global plan or filter whole tool outputs lose utility because the predicted full trajectory almost never matches the executed one once environmental feedback arrives.Following instructions hidden in data · reviewed by AI
Notes
Drafted from the paper by a model and filed unreviewed. Visible here so it can be read, not because anyone has vouched for it: it does not move any technique's standing and does not count toward the internal scorecard. Drafted confidence: medium. Falsifier as drafted: Traces from the same environment and protocol showing frontier agents allocating a substantially larger share of calls to global-state monitoring, raising monitoring as the endgame nears, and executing the large majority of their stated near-term commitments within ten turns.