GEM-Bench
A benchmark for evaluating the effectiveness of native advertising in LLMs, covering three typical application scenarios: question answering, recommendation and dialogue. It scores along dimensions including advertising relevance, answer naturalness and overall score.
Timeline (last 90 days)
The paper evaluates on GEM-Bench covering three application scenarios, and AdSteer improves the overall score by an average of 14.1% over the strongest baseline.
GEM-Bench is the benchmark used in the paper to evaluate the effectiveness of LLM-native advertising, covering three typical application scenarios (such as question answering, recommendation, and dialogue) and scoring on dimensions such as ad relevance, response naturalness, and overall score.
The current experiments are based on the GEM-Bench benchmark, and factors such as user tolerance and the diversity of advertiser demands in real product environments still require further verification.
All Facts (3)
The paper evaluates on GEM-Bench covering three application scenarios, and AdSteer improves the overall score by an average of 14.1% over the strongest baseline.
50%UnverifiedGEM-Bench is the benchmark used in the paper to evaluate the effectiveness of LLM-native advertising, covering three typical application scenarios (such as question answering, recommendation, and dialogue) and scoring on dimensions such as ad relevance, response naturalness, and overall score.
50%UnverifiedThe current experiments are based on the GEM-Bench benchmark, and factors such as user tolerance and the diversity of advertiser demands in real product environments still require further verification.
50%