CARE-PO: Controllable Tool-Call Efficiency for LLM Agents
Organizations: Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) · Shenzhen University · China Southern Power Grid Company Limited
Abstract
External tool calls are a major running cost of an LLM agent, while its training objectives align almost exclusively with task completion. In a step-level diagnosis of Qwen3.5 trajectories at two model sizes on GAIA, a long-horizon tool-use benchmark, we find that most contain removable calls, at a similar rate at both sizes. Some methods add an efficiency reward to the objective to reduce the call count, but introduce problems of their own: the anchor they price calls against is computed from the policy's own rollouts (e.g., the group's shortest correct rollout) and so moves with the policy; training can drift toward a collapsed policy that issues no tool call at all, a degeneracy seen in published reports and in our 27B baseline; and a training run delivers a single operating point, so the user cannot trade efficiency against capability. We present CARE-PO (Controllable Anchored Relative Efficiency Policy Optimization) to mitigate these problems. The efficiency reward prices calls against a reference budget anchored to the task's own difficulty and fixed before training. Injecting a successful trajectory from the training data into each online GRPO rollout group and charging for evidence a trajectory never gathers ease the drift toward a collapsed call count. Shaping the cost curve by an effort tag in the prompt yields an efficiency and a capability operating point from a single training run. On GAIA, five of the six operating points across three model sizes improve on their base model in success rate and call count at once. On four held-out multi-hop benchmarks the result is size-dependent: at 27B and 4B success rate is held or raised at a lower call count, while at 9B both operating points trade success rate for fewer calls. The call count remains well above zero throughout training at all three sizes.
Figures & tables
| Method | SR | AC | HV |
| Qwen3.5-9B | |||
| Vanilla ReAct | 40.2 0.7 | 8.26 0.14 | 0.236 0.005 |
| Budget-Prompt | 35.4 0.3 4.8 | 8.50 0.05 2.9% | 0.203 0.001 14.0% |
| DEPO | 34.7 0.9 5.5 | 7.54 0.05 8.7% | 0.217 0.005 8.1% |
| OTC-PO | 40.6 1.2 0.4 | 5.01 0.13 39.3% | 0.304 0.012 28.8% |
| CARE-PO (low) | 39.8 2.3 0.4 | 3.42 0.30 58.6% | 0.330 0.015 39.8% |
| 2Wiki | HotpotQA | MuSiQue | Bamboogle | |||||||||
| Method | SR | AC | HV | SR | AC | HV | SR | AC | HV | SR | AC | HV |
| Qwen3.5-9B | ||||||||||||
| Vanilla ReAct | 75.8 | 5.64 | 0.544 | 48.6 | 4.87 | 0.368 | 29.0 | 6.59 | 0.194 | 76.0 | 4.05 | 0.606 |
| OTC-PO | 70.6 | 2.24 | 0.627 | 48.0 | 1.59 | 0.442 | 25.6 | 2.23 | 0.227 | 68.8 | 1.38 | 0.640 |
| CARE-PO (low) | 63.2 | 1.94 | 0.571 | 41.4 | 1.09 | 0.391 | 21.4 | 1.55 | 0.197 | 43.2 | 0.50 | 0.421 |
| CARE-PO (high) | 70.4 | 2.19 | 0.627 | 46.2 | 1.51 | 0.427 | 25.2 | 2.10 | 0.226 | 60.8 | 1.20 | 0.572 |
| Configuration | SR | AC | HV | Note |
| Trained at , low tag | 39.2 | 4.07 | 0.312 | same price at every tier |
| Trained at , high tag | 40.6 | 4.91 | 0.306 | as above |
| Same weights, no tag at inference | 39.8 | 4.29 | 0.313 | the unconditioned tier from training |
| Untrained base + low tag | 37.0 | 7.01 | 0.240 | no training, prompt only |
| Untrained base + high tag | 40.6 | 7.77 | 0.248 | no training, prompt only |
| CARE-PO (low) | 39.8 | 3.42 | 0.330 | efficiency operating point |
Appendix figures & tables4 assets
Supplementary material from the paper’s appendix.
Appendix
| Method | AC | Turns | Generated | Observations | Context processed |
| Vanilla ReAct | 8.26 | 9.1 | 1.91k | 13.1k | 104.4k |
| OTC-PO | 5.01 | 6.0 | 1.41k | 8.90k | 53.2k |
| CARE-PO (low) | 3.42 | 4.4 | 1.04k | 6.54k | 34.6k |
| CARE-PO (high) | 4.44 | 5.4 | 1.24k | 8.11k | 45.3k |
LLM Agent Training
Large language model agent training teaches models to plan, use tools, and learn from multi-step interaction with environments.
See all 623 papers