ConflictQA - KG Conflict (Complementary): leaderboard
Metric: Exact match (%) on 430 complementary (answers need both sources) questions when the provided evidence contains conflicting knowledge-graph triples (TripleConf); zero-shot with a conflict-aware prompt, the mean of two evidence orders; exact match of the generated answer set; higher is better. Source: arxiv.org. Saturation forecast: Around 2029. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 Mini | 37.67 |
| 2 | DeepSeek V3.2 (Thinking) | 36.98 |
| 3 | GPT-4o | 33.72 |
| 4 | GPT-5.1 | 33.72 |
| 5 | Llama 3.1 70B Instruct | 31.16 |
| 6 | GPT-3.5 Turbo (0125) | 30.7 |
| 7 | Mistral Large 3 | 29.77 |
| 8 | DeepSeek V3.2 (Non-reasoning) | 26.74 |
| 9 | Llama 3.1 8B Instruct | 17.67 |
| 10 | Qwen 3 8B | 8.14 |
Interactive version: theaggregate.ai/benchmark?slug=conflictqa-kg-conflict-complementary · How It Works · Data refreshed daily, snapshot 2026-10-07.