# Long-horizon tool use

3 related articles

Subscribe to updates

arXiv Nov 3, 2026
ToolSpace rebuilds agent training environments from real tool-call traces, lifting a 35B model's average score from 23.0 to 41.0 across seven benchmarks
ToolSpace reverses the order in which agent training environments are synthesized: it first mines data, state and retry dependencies between adjacent calls in execution traces from real MCP services and open agent corpora using deterministic rules, clusters the resulting weighted tool-composition graph into environments, synthesizes databases only where state is mutated, and then runs state-aware random walks with rollback to decompile replayable chains into natural-language tasks with assertion-based checks. Across seven long-horizon tool-use benchmarks, Qwen3.6-35B-A3B rises from 23.0 to 41.0 average score, or 43.2 with agent RL, and GLM-5.2 rises from 44.6 to 56.6.
ToolSpace reverses the order in which agent training environments are synthesized: it first mines data, state and retry dependencies between adjacent calls in execution traces from real MCP services and open agent corpora using deterministic rules, clusters the resulting weighted tool-composition graph into environments, synthesizes databases only where state is mutated, and then runs state-aware random walks with rollback to decompile replayable chains into natural-language tasks with assertion-based checks. Across seven long-horizon tool-use benchmarks, Qwen3.6-35B-A3B rises from 23.0 to 41.0 average score, or 43.2 with agent RL, and GLM-5.2 rises from 44.6 to 56.6.
ToolSpace Tool-Centric Environment Synthesis Long-horizon tool use Model Context Protocol AutomationBench Pro
arXiv Oct 5, 2026
CITA trains a comparative inference model with a Bayesian tool-graph simulator and LLM comparison judgments, consistently improving Tool F1 and task success across three tool-use benchmarks and multiple backbone LLMs
The work proposes Comparative Inference for Tool-use Agents (CITA), which trains a Comparative Inference Model (CIM) from paired signals combining observed tool behavior, scalable supervision from a Bayesian tool-graph simulator, and semantic judgments from LLM-based comparison, so that the model estimates how likely a possible next tool invocation is to support final task success under the current context; across three tool-use benchmarks and multiple backbone LLMs, CITA consistently improves Tool F1 and task success, and additional analysis shows CIM learns accurate step-level value estimates for comparative tool choices.
The work proposes Comparative Inference for Tool-use Agents (CITA), which trains a Comparative Inference Model (CIM) from paired signals combining observed tool behavior, scalable supervision from a Bayesian tool-graph simulator, and semantic judgments from LLM-based comparison, so that the model estimates how likely a possible next tool invocation is to support final task success under the current context; across three tool-use benchmarks and multiple backbone LLMs, CITA consistently improves Tool F1 and task success, and additional analysis shows CIM learns accurate step-level value estimates for comparative tool choices.
Shandong Jianzhu University Southeast University
Yu Li（First author）
Comparative Inference for Tool-use Agents Comparative Inference Model Long-horizon tool use
arXiv Oct 6, 2026
CIPO trains LLM agents with counterfactual branches to adaptively pick tool granularity, raising success and decision compression across three benchmarks and two backbones
The work proposes CIPO: it first mines executable composite skills from successful tool-use trajectories via budget-constrained BPE-style merging, then augments GRPO by branching at the first eligible granularity decision of each base rollout and using the paired outcome difference as a bounded supplementary reward; across TOOLATHLON, TOUCAN, and TRAJECT-Bench with Qwen2.5-7B and Llama3.1-8B, CIPO attains the highest Tool F1 and TSR in all six settings with an average Comp./Raw of 0.845 versus 0.912 for GRPO + Skills, and the gain does not come from simply invoking skills more often.
The work proposes CIPO: it first mines executable composite skills from successful tool-use trajectories via budget-constrained BPE-style merging, then augments GRPO by branching at the first eligible granularity decision of each base rollout and using the paired outcome difference as a bounded supplementary reward; across TOOLATHLON, TOUCAN, and TRAJECT-Bench with Qwen2.5-7B and Llama3.1-8B, CIPO attains the highest Tool F1 and TSR in all six settings with an average Comp./Raw of 0.845 versus 0.912 for GRPO + Skills, and the gain does not come from simply invoking skills more often.
Huawei Technologies (China)KTH Royal Institute of Technology Southeast University
Yu Li（First author）
CIPO Adaptive tool granularity selection Executable composite skills Group-Relative Policy Optimization Long-horizon tool use

You’re all caught up

ToolSpace rebuilds agent training environments from real tool-call traces, lifting a 35B model's average score from 23.0 to 41.0 across seven benchmarks

CITA trains a comparative inference model with a Bayesian tool-graph simulator and LLM comparison judgments, consistently improving Tool F1 and task success across three tool-use benchmarks and multiple backbone LLMs

CIPO trains LLM agents with counterfactual branches to adaptively pick tool granularity, raising success and decision compression across three benchmarks and two backbones