Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation
This work introduces a two-level framework that distills LLM reasoning into an efficient non-generative student and adapts its decision boundary to product-type-specific trade-up criteria.
Si-Liang Liu, Mohammadhasan Ghasemi, Sapan Patel et al.
· 0 citations