open-source-llm-agents-driving-ros-robot-having-model-emit CLM-0462For open-source LLM agents driving a ROS robot, having the model emit a pseudo-code tool-call plan up front, storing it in a persistent scratchpad, and then restricting each tool invocation to the plan's next step raises task success most on the longest multi-step tasks and for the smaller open models (GPT-OSS-20B on the hardest task rose from 35% with a persistent plan alone to 55% with the constraint), while a strong proprietary model gains little.
Capability: Completing tasks that take many steps or many hours
Observed on
ROS-Agent/ROSA-style agentic loop with about eight custom robot tools; three task tiers; 20 trials per cell in Gazebo simulation plus 40 hardware trials on one task; open models GP.
Sources
- Validated by: Four-configuration ablation (prompt only, transient plan, persistent plan, persistent plan plus execution constraint) over three task tiers and four models, 20 trials each; 40 hardware trials comparing repeated tool calls (19.0% to 6.1%), invalid calls (21.7% to 8.1%) and execution time against vanilla ROSA; and a 40-trial failure-mode comparison against re-implemented ReAct (17/40 success), CaP (20/40), ROSA (23/40) and MetaTool (30/40).. [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 tool-integrated LLM agents where the attack works by adding attacker-controlled tools, overwriting the agent's visible toolset with the saved user toolset immediately before planning drives attack success to near zero across direct and indirect injection, memory poisoning and backdoor attacks, but its effect on benign task success is model-dependent — raising it for 7-9B open models and lowering it for GPT-3.5/GPT-4.Following instructions hidden in data · reviewed by AI
- For LLM-driven robot manipulation where every tool call is schema-validated and every motion kinematically and collision-checked before execution, execution-time recovery mechanisms (retrying a failed plan with an alternative candidate, resetting to a collision-checked safe pose between stages) contribute more to end-to-end task success than canonical action sequencing, because most residual failures are transient planning or IK failures rather than wrong orderings.Turning plans into physical action · reviewed by AI
- For LLM-guided narrow-space parking where an optimal-control solver supplies pass/fail evidence, growing a decision knowledge base offline from accumulated solver and validation failures — retaining an entry only when re-planning then yields a feasible trajectory — raises planning success from 85.1% to 94.7% and closed-loop success from 74.6% to 84.2% over an empty knowledge base, with model weights frozen.Improving its own scaffolding from failures · reviewed by AI
- 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 agents that must compose and then execute domain tool workflows, training separate adapters for the two roles splits the gains cleanly: supervised fine-tuning of the planner on reference tool sequences mainly improves workflow-order fidelity, while node-level RL on argument-match rewards mainly improves tool-argument grounding and final-answer quality, and combining both beats either on a 9B backbone.Planning a sequence of actions before acting · 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: low. Falsifier as drafted: An ablation on the same pipeline where adding the execution constraint to a persistent plan does not raise success on the longest tasks, or where gains are as large for strong proprietary models as for small open ones, or where repeated/invalid tool-call rates are unchanged.