Researchers have introduced PhoneValue-Bench, a new benchmark designed to evaluate AI models for smartphone recommerce inspection. This framework addresses the need for assessing payout-relevant scorecards, which include both cosmetic and operational ground truth, a gap not covered by existing defect localization or attribute-focused resources. The benchmark comprises the PhoneValue-Corpus for training on a large dataset of trade-in orders and PV-Rank for leaderboard evaluation on a smaller, audited set of orders, aiming to provide a more comprehensive assessment than simple price prediction. AI
IMPACT This benchmark could improve the accuracy and efficiency of AI models used in the growing smartphone recommerce market.
RANK_REASON The cluster describes a new benchmark and corpus for evaluating AI models in a specific domain (smartphone recommerce inspection), presented in an academic paper. [lever_c_demoted from research: ic=1 ai=1.0]
- CardScore
- Flat Vision-Language Model
- Leadgate
- Operational Fill
- PhoneValue-Bench
- PhoneValue-Corpus
- PV-Rank
- ValueAgent
- Zihan Deng
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →