Teacher-Guided Tool Learning for Small Language Models via Bidirectional KL Constraints
cs.LG, cs.CL
Submitted: 2026-10-13
Updated: 2026-10-13
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
- Training Verifiers to Solve Math Word Problems
- Teaching LLMs to Learn Tool Trialing and Execution through Environment Interaction
- Distill Where You Fail: Recovering Learning Signals of Negative RL-Groups from Adaptive Teacher Guidance
- Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
- Student-Centered Distillation Narrows the Agentic Gap Between Small and Large LLMs
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- KDRL: Post-Training Reasoning LLMs via Unified Knowledge Distillation and Reinforcement Learning
- Qwen2.5 Technical Report
- SOD: Step-wise On-policy Distillation for Small Language Model Agents
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