TRACK - Multi-Hop QA (MeLLo): leaderboard
Knowledge-conflict reasoning in TRACK (Testing Reasoning Amid Conflicting Knowledge; Feng et al., EACL 2026): 500 multi-hop questions over Wikidata fact chains, where the needed facts contradict or extend what the model learned in pre-training, answered through MeLLo, which keeps the facts in an external memory and retrieves them by question decomposition. Holistic pass in percent (a correct final answer and reasoning that entails every required fact), higher is better.
Metric: Holistic pass (%; multi-hop questions over Wikidata fact chains, 500 items; the final answer is correct and the reasoning entails every required fact; answered through MeLLo, which keeps the facts in an external memory and retrieves them by question decomposition). Source: arxiv.org. Saturation forecast: Around July 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 8B | 10.6 | #1838 |
| 2 | Qwen 3 4B | 7.9 | #2297 |
| 3 | Qwen 3 1.7B | 5.5 | #5551 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=track-multi-hop-qa-mello · How It Works · Data refreshed daily, snapshot 2026-11-12.