Tag
The paper introduces a two-stage LLM pipeline using fine-tuned Qwen3-4B with Hyper-Parallel Decoding to extract purchase-discriminative attributes in e-commerce, achieving 85% accuracy with 92% cost reduction.
This paper presents SynthAVE, a large-scale human-validated benchmark for attribute value extraction in e-commerce, using a multi-LLM arena framework with 21 judge configurations to validate synthetic labels efficiently and cost-effectively while maintaining quality parity with human review.