ConflictQA - KG Conflict (Non-Complementary): leaderboard

ConflictQA (2026) asks questions whose evidence, knowledge-graph triples plus Wikipedia text, contains a contradiction injected into one source; the model is told the sources may conflict and answers zero-shot, scored by exact match of the answer set and averaged over two evidence orders. This board covers single-hop questions with falsified triples beside correct text. Exact match in percent; higher is better.

Metric: Exact match (%) on 802 single-hop (non-complementary) 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. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1O3 Mini38.03#546
2GPT-3.5 Turbo (0125)36.41#3550
3DeepSeek V3.2Thinking33.23#419
4GPT-4o32.67#751
5Llama 3.1 70B Instruct32.67#1571
6ConflictQA Qwen3-30B-A3B-Thinking (checkpoint unspecified)32.61
7GPT-5.132.05#240
8Mistral Large 329.3#832
9Llama 3.1 8B Instruct19.2#4600
10open-mistral-7B13.47#7710
11Qwen 3 8B7.61#1873

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=conflictqa-kg-conflict-non-complementary · How It Works · Data refreshed daily, snapshot 2026-11-21.