Question Answering on ConflictQA benchmark leaderboard · SOTA2 Research

Question Answering on ConflictQA

53.8AccuracyCR-DPO

Evaluation Results

Method	Links
CR-DPO2026.05		53.8	94.9	81.6	88.2
SABER2026.05		53.5	97	85	91
SABER2026.05		49.6	99.3	90	94.6
Implicit-SCR2026.05		47.8	94.5	36	65.3
CR-DPO2026.05		47.8	98.5	61	79.8
Implicit-SCR2026.05		47.3	89.5	36.1	62.8
TPC2026.05		47	92.7	34.7	63.7
SABER2026.05		46.9	97.9	82.9	90.4
Vanilla RAG2026.05		46.8	100	7.5	53.7
Vanilla RAG2026.05		44.5	100	7	53.5
TPC2026.05		44.3	96	27	61.5
SABER2026.05		44	98	75.4	86.7
Implicit-SCR2026.05		43.4	92.9	19.3	56.1
TACS-LR2026.05		42.9	90.1	12.2	51.2
Vanilla RAG2026.05		42.1	100	7	53.5
Vanilla RAG2026.05		42	100	5.4	52.7
TACS-LR2026.05		41.9	84	13	48.5
TPC2026.05		41.6	96.7	40.4	68.5
TPC2026.05		41.4	92.4	37.8	65.1
Implicit-SCR2026.05		41.3	88.1	20.3	54.2
Explicit-SCR2026.05		40.9	73.9	52.4	63.1
CR-DPO2026.05		39.8	79.2	85.6	82.4
TACS-LR2026.05		39.5	90.9	3.5	47.2
CR-DPO2026.05		38.6	92.2	31.6	61.9
Internal-Eval2026.05		38.2	72.1	63	67.6
Internal-Eval2026.05		37.2	74.1	65	69.5
TACS-LR2026.05		36.7	82.9	6.2	44.6
Explicit-SCR2026.05		34.8	70.7	30	50.3
Internal-Eval2026.05		31.9	37.6	72.1	54.9
Explicit-SCR2026.05		31.8	78.4	14	46.2
Context-Eval2026.05		31.7	36.6	79.6	58.1
Context-Eval2026.05		31	55.9	50	53
Internal-Eval2026.05		30.6	64.9	57.9	61.4
Context-Eval2026.05		29.5	51.9	68.1	60
Explicit-SCR2026.05		27.6	54.6	68.5	61.6
Closed-Book2026.05		22.7	3	100	51.5
R-Tuning2026.05		15.9	4.4	70.9	37.7
Closed-Book2026.05		14.6	1.3	100	50.6
Closed-Book2026.05		13.9	0.8	100	50.4
Context-Eval2026.05		12.3	13.8	70.2	42
R-Tuning2026.05		10.5	2.6	28.3	15.5
R-Tuning2026.05		10.4	6.4	46.2	26.3
Closed-Book2026.05		9.1	1.1	100	50.6
R-Tuning2026.05		0.4	0	0.9	0.5