Trajectory-Refined Distillation

arXiv:2606.08432 · cs.AI · Submitted 2026-06-07 · Read on arXiv

cs.AI

Submitted: 2026-06-07

Updated: 2026-10-12

Code: https://github.com/louieworth/trd

Terminology

• Distillation — 5× in this paper · explained in 2 episodes
• Trajectory-Refined Distillation — 2× in this paper
• Accuracy — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• On-Policy Distillation — 1× in this paper · explained in 3 episodes
• On-policy Self-Distillation (OPSD) — 1× in this paper · explained in 2 episodes
• Self-Distillation — 1× in this paper · explained in 7 episodes

Sources

• AMO-Bench: Large Language Models Still Struggle in High School Math Competitions
• Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
• Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes
• Distilling the Knowledge in a Neural Network
• Reinforcement Learning via Self-Distillation
• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
• Stable On-Policy Distillation through Adaptive Target Reformulation
• Entropy-Aware On-Policy Distillation of Language Models
• Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
• Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
• TACO: Topics in Algorithmic COde generation dataset
• Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
• Prefix Teach, Suffix Fade: Local Teachability Collapse in Strong-to-Weak On-Policy Distillation
• Demystifying OPD: Length Inflation and Stabilization Strategies for Large Language Models
• Privileged Information Distillation for Language Models
• Self-Distillation Enables Continual Learning
• HybridFlow: A Flexible and Efficient RLHF Framework
• Experiential Reinforcement Learning
• A Survey of On-Policy Distillation for Large Language Models
• GATES: Self-Distillation under Privileged Context with Consensus Gating

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems