Dandan Tu, Rui Liu, Ziru Liu, Hanbo Li, Jinpeng Chen, Suiyun Zhang, Chenyang Zhang, Wu Zhang, Shi Wu, Xinyu Fu, Cheng Gong, Zichen Tian
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. We have no repository linked to it either. That is a statement about our graph, not about the paper.
Developing multi-turn interactive tool-use agents is challenging because real-world user needs are often complex and ambiguous, yet agents must execute deterministic actions to satisfy them. To address this gap, we introduce CoVe (Constraint-Verification), a post-training data synthesis framework designed for training interactive tool-use agents while ensuring both data complexity and correctness. CoVe begins by defining explicit task constraints, which serve a dual role: they guide the generation of complex trajectories and act as deterministic verifiers for assessing trajectory quality. This enables the creation of high-quality training trajectories for supervised fine-tuning (SFT) and the derivation of accurate reward signals for reinforcement learning (RL). Our evaluation on the challenging τ 2 -bench benchmark demonstrates the effectiveness of the framework. Notably, our compact CoVe-4B model achieves success rates of 43.0% and 59.4% in the Airline and Retail domains, respectively; its overall performance significantly outperforms strong baselines of similar scale and remains competitive with models up to 17× its size. These results indicate that CoVe provides an effective and efficient pathway for synthesizing training data for state-of-the-art interactive tool-use agents. To support future research, we open-source our code, trained model, and the full set of 12K high-quality trajectories used for training.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2603.01940")
get_code_for_paper("2603.01940")
have("2603.01940")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2603.01940.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2603.01940)
Connect an agent — have() is free.