ConflictQA - KG Conflict (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 questions that need both sources, with falsified triples added. Exact match in percent; higher is better.

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 June 2028. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1O3 Mini37.67#546
2DeepSeek V3.2Thinking36.98#419
3ConflictQA Qwen3-30B-A3B-Thinking (checkpoint unspecified)34.88
4GPT-4o33.72#751
5GPT-5.133.72#240
6Llama 3.1 70B Instruct31.16#1571
7GPT-3.5 Turbo (0125)30.7#3550
8Mistral Large 329.77#832
9Llama 3.1 8B Instruct17.67#4600
10Qwen 3 8B8.14#1873
11open-mistral-7B6.51#7710

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-complementary · How It Works · Data refreshed daily, snapshot 2026-11-21.