drone-s-action-space-declared-only-prompt-model-must CLM-0379When a drone's action space is declared only in the prompt and the model must itself declare arrival, small open multimodal models (2B-9B) enter the 5 m success radius as often as or more often than frontier models yet lose most episodes by declaring arrival too early or never declaring, so the gap at small scale is protocol adherence rather than navigation.
Capability: Turning plans into physical action
Observed on
Simulated AirSim drone episodes, monocular RGB only, four prompt-declared actions (go, rotate, think, finished), no fine-tuning or function-calling schema; 20 episodes per task cel.
Sources
- Validated by: Scored 9 models over 80 single-drone and 20 four-drone episodes; comparing success rate to oracle-success (trajectory entered the radius) exposes a declaration gap invisible under threshold-based termination (e.g. Qwen3.5-9B 90% oracle vs 35% success), and 61 of 337 never-declaring failures had entered the radius.. Measured on 100 simulated episodes, 20 per cell, so each point is 5 percentage points and the authors' own variance audit gives 9-13 point per-cell standard deviations; the declaration gap (90% vs 35% oracle-vs-success for Qwen3.5-9B; [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
- For long-horizon embodied task planning where a symbolic model defines action preconditions and effects, correcting a violated plan by edits the action model itself implies (inserting the missing navigate or open) raises plan success more per LLM call than returning the violation to the LLM for replanning, and the gain persists for frontier models as well as compact ones.Planning a sequence of actions before acting · reviewed by AI
- For LLM-driven multi-UAV swarm tasks in simulation, binding one agent to each vehicle with a role-conditioned subset of the skill catalog and directed peer messages yields higher episode returns than a central planner that sees pooled team state and the full skill list, and ablations attribute most of the gain to the skill-subset restriction and the peer reports rather than to the experience-based reward correction.Turning plans into physical action · reviewed by AI
- In closed-book multi-hop QA the accuracy gain from decomposed prompting shrinks to little in frontier models, but answers still change between task-equivalent prompt regimes (agreement only about 70%, and about 60% on MuSiQue), and because stable facts survive re-prompting while hallucinations do not, disagreement across regimes detects errors better than standard uncertainty baselines on F1 and AUROC without retrieval or training.Recalling facts it was trained on · reviewed by AI
- On cross-system urban tasks where several services must be chained under matched tools and budgets, the task-success gap between orchestration strategies is not explained by required-tool coverage, call ordering, or asking for missing inputs — those are near-saturated across strategies — but by whether the agent goes beyond the minimal toolchain to refine coarse results and verify each candidate against the stated constraints before answering.Using the tools it is given · reviewed by AI
- For training-free MLLM-driven navigation (instruction-following and object-goal) in simulated indoor environments, gating each planner-proposed action on evidential support, geometric reachability and subgoal consistency before dispatch — and refusing a stop request unless semantic, geometric and progress checks all pass — raises success rate over letting the planner act directly, but the stop check buys completion at some cost in path efficiency.Turning plans into physical action · 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: A run in which small open models' oracle-success and success rates track each other as closely as frontier models' do, with failures attributable to not reaching the target rather than to mistimed or missing declarations.