Davinci-Tool: Scaling Agentic Tool-Use Supervision with Verifiable Executable Environments
cs.LG, cs.AI, cs.CL
Submitted: 2026-10-12
Updated: 2026-10-12
Comments: 27 pages, 4 figures, 13 tables. Code: https://github.com/MassLab-SII/Davinci-Tool
Code: https://github.com/MassLab-SII/Davinci-Tool
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Synthesize and Reward -- Reinforcement Learning for Multi-Step Tool Use in Live Environments
- MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- Training Deep Nets with Sublinear Memory Cost
- Decoupled Weight Decay Regularization
- Execution-First Synthetic Tool-Use Trace Generation for LLM Agents
- Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents
- Kimi K2: Open Agentic Intelligence
- ASTRA: Automated Synthesis of agentic Trajectories and Reinforcement Arenas
- Towards Self-Improvement of LLMs via MCTS: Leveraging Stepwise Knowledge with Curriculum Preference Learning
- AutoWebWorld: Synthesizing Infinite Verifiable Web Environments via Finite State Machines
- Monte Carlo Tree Search Boosts Reasoning via Iterative Preference Learning
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments
- Qwen3 Technical Report
- Automated Skill Discovery for Language Agents through Exploration and Iterative Feedback
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- AgentEvolver: Towards Efficient Self-Evolving Agent System
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks