SocialPersona - Dialogue (Timeline): leaderboard

Metric: Dialogue score (0 to 5): GPT-5.5 judge mean of interest coverage, concreteness and fluency of responses to stable-interest recommendation and recent-interest exploration requests, judged against the gold profile; the response is written with the raw timelines of 100 social-media users given as post text, image captions and timestamps (up to 200 posts) in context; higher is better. Source: arxiv.org. Saturation forecast: Around March 2027. 6 models tracked.

Top models

#ModelScore
1GPT-5.43.57
2Qwen 3 VL 8B Instruct3.3
3GPT-4o Mini3.13
4Gemini 2.5 Flash2.96
5Qwen 3.5 35B A3B2.96
6Qwen 2.5 VL 7B Instruct2.55

Interactive version: theaggregate.ai/benchmark?slug=socialpersona-dialogue-timeline · How It Works · Data refreshed daily, snapshot 2026-09-29.