Learn from Whoever Is Right: Answer-Verified Multi-Teacher Distillation for Multi-Domain LLMs

arXiv:2609.02548 · cs.LG, cs.AI · Submitted 2026-09-02 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-02

Updated: 2026-09-02

Code: https://github.com/hexixiang/MT-SDPO

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Distillation — 41× in this paper · explained in 2 episodes
• Verifier — 29× in this paper · explained in 2 episodes
• Accuracy — 26× in this paper · explained in 1 episode
• Verification — 23× in this paper · explained in 2 episodes
• Aggregation — 15× in this paper · explained in 1 episode
• GRPO — 14× in this paper · explained in 6 episodes
• Self-Distillation — 7× in this paper · explained in 7 episodes
• On-Policy Distillation — 6× in this paper · explained in 3 episodes
• Ablation — 5× in this paper · explained in 1 episode
• Reinforcement Learning — 5× in this paper · explained in 4 episodes
• Knowledge Distillation — 4× in this paper · explained in 11 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Ablation Studies — 1× in this paper · explained in 1 episode
• Exponential Moving Average (EMA) — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Large Language Model (LLM) — 1× in this paper · explained in 3 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• Advancing General-Purpose Reasoning Models with Modular Gradient Surgery
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
• SciKnowEval: Evaluating Multi-level Scientific Knowledge of Large Language Models
• The Llama 3 Herd of Models
• Distilling the Knowledge in a Neural Network
• Reinforcement Learning via Self-Distillation
• Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
• Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR
• Tulu 3: Pushing Frontiers in Open Language Model Post-Training
• MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training
• WebGPT: Browser-assisted question-answering with human feedback
• Proximal Policy Optimization Algorithms
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Expanding the Capabilities of Reinforcement Learning via Text Feedback
• MiMo-V2-Flash Technical Report
• Olmo 3
• Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
• Qwen3 Technical Report
• Disentangling Task Conflicts in Multi-Task LoRA via Orthogonal Gradient Projection
• Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation

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: Local-Consistency Optimisation for Bayesian Neural Networks
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks