MediLongChat - Cross-Dialogue Reasoning: leaderboard
MediLongChat (Hu et al., South-Central Minzu University and Singapore Management University, 2026) builds synthetic longitudinal doctor-patient dialogue histories for 80 patients from knowledge-guided records, about 50,000 tokens over 18 visits per patient. Cross-dialogue reasoning asks about relations that only several visits together settle, such as the order of events, the time between two illnesses, recurrence or a change of therapy, given the whole history; the score is token-level F1 against the reference answer in percent, higher is better.
Metric: Token F1 (%) of short answers linking facts across a patient's visits. Source: arxiv.org. Saturation forecast: Around April 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Mini | 24.25 | #1627 |
| 2 | GPT-4.1 Mini | 18.37 | #775 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=medilongchat-cross-dialogue-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-12.