Share your thoughts, 1 month free Claude Pro on usSee more

Multimodal Reward Modeling on VL-RewardBench

83.5Accuracy

DT2IT-MRM (Qwen3-VL)

Updated 1mo ago

Evaluation Results

Method
DT2IT-MRM (Qwen3-VL) 2026.04	83.5	77	63	91.3	76.7
DT2IT-MRM 2026.04	83.5	-	-	-	-
DT2IT-MRM (Qwen2.5-VL) 2026.04	82.5	75.8	63.5	91.3	72.6
BaseReward (Qwen2.5-VL) 2026.04	82.2	80.9	68.6	92.2	81.8
BaseReward 2026.04	82.2	-	-	-	-
MSRL + voting@16 2026.03	77.5	-	76.5	79.3	76.8
EGT 2026.02	77.15	-	-	-	-
MSRL 2026.03	75.9	-	75.4	78.2	74.2
MR. Judge-7B-SFT-RL 2026.04	75.5	71.1	68.7	83.2	61.4
Proxy-GRM-RL 2026.03	75.22	73.93	-	-	-
Gemini-2.5-Pro 2026.02	74.9	72.9	59.1	85.2	74.4
Unified-Reward-Think 2026.03	73.8	72.3	-	-	-
Gemini-3-Pro 2026.03	73.4	-	54.6	78.9	86.7
Proxy-GRM-RL 2026.03	73.38	72.38	-	-	-
Skywork-VL Reward 2026.04	73.1	69	66	80	61
Skywork-VL Reward 2026.04	73.1	-	-	-	-
VL-MDR 2026.04	73.06	71.84	71.27	75.17	69.09
UnifiedReward-Think* 2026.04	73	65.4	51.4	82.9	61.8
Skywork-VL-Reward 2026.04	72.98	68.82	65.75	79.84	60.88
R1-Reward 2026.02	72.89	-	-	-	-
UnifiedReward-think (w/ CoT) 2026.03	72.2	-	77.9	72.7	66
Proxy-GRM-RL 2026.03	72.17	71.18	-	-	-
Omni-RewardModel-BT* 2026.04	72	64.8	51.4	81.4	61.5
R1-Reward 2026.03	71.92	71.44	-	-	-
Gemini-2.5-Flash 2026.02	71.9	72	58	77	81
UnifiedReward-Think 2026.04	71.45	71.82	77.35	72.5	65.62
R1-Reward 2026.03	71.4	-	63.8	85.7	64.8
MSRL + voting@16 2026.03	71.1	-	63.5	82.2	67.6
GPT-5-mini 2026.03	70.9	-	60.2	76.5	76
UnifiedReward 2026.03	70.8	-	76.5	70.5	65.4
GPT-5.2* 2026.04	70.2	67	52.5	71.8	76.7
GPT-5.2 2026.04	70.2	-	-	-	-
MSRL 2026.03	69.7	-	62.4	80.5	66.3
Proxy-GRM-SFT 2026.03	69.53	69.18	-	-	-
Gemini-1.5-Pro 2026.04	67.2	62.5	50.8	72.5	64.2
Gemini-1.5-Pro (2024-09-24) 2026.04	67.2	62.5	50.8	72.5	64.2
Generative MRM 2026.03	66.6	-	57.2	74.5	68.2
Claude-3.7-Sonnet 2026.03	66.5	-	68.1	70.7	60.8
Claude-3.7-Sonnet (2025-02-24) 2026.02	66.31	-	-	-	-
Claude-3.7-Sonnet 2026.03	66.31	66.53	-	-	-
Claude 3.7 Sonnet 2026.04	66.31	66.53	68.08	70.7	60.81
Claude-3.7-Sonnet 2026.04	66.3	66.5	68.1	70.7	60.8
IXC-2.5-Reward 2026.04	66.16	68.55	80.11	65.29	60.25
Unified-Reward-SFT 2026.03	66.1	66.5	-	-	-
GPT-4o (2024-08-06) 2026.02	65.8	-	-	-	-
IXC-2.5-Reward 2026.02	65.8	-	-	-	-
GPT-4o 2026.02	65.8	62.4	49.6	67.6	70.5
GPT-4o-(2024-08-06) 2026.03	65.8	62.4	-	-	-
IXC-2.5-Reward 2026.03	65.8	70	-	-	-
GPT-4o 2026.04	65.8	62.4	49.1	67.6	70.5
GPT-4o (2024-08-06) 2026.04	65.8	62.4	49.1	67.6	70.5
IXC-2.5-Reward 2026.04	65.8	70	84.7	62.5	62.9
Discriminative MRM 2026.03	64.3	-	54.8	75.6	62.4
InternVL3-78B 2026.04	63.3	61.6	67.8	52.5	64.5
UnifiedReward* 2026.04	63	58.8	45.9	66.6	64
UnifiedReward 2026.04	62.79	66.61	76.24	58.61	64.98
Gemini-1.5-Pro 2026.03	62.5	-	50.8	72.5	64.2
GPT-4o 2026.03	62.4	-	49.1	67.6	70.5
Discriminative MRM 2026.03	61.2	-	54.4	66.8	62.4
Generative MRM 2026.03	60.5	-	56.6	65.4	59.4
InternVL3-78B 2026.04	57.98	62.15	69.61	52.47	64.35
Gemini-1.5-Flash 2026.04	57.6	55.3	47.8	59.6	58.4
PhyCritic-7B 2026.02	57.3	54.9	45.3	58.6	60.9
InternVL3-8B 2026.04	57	55.6	60.6	44	62.3
Llama-3.2-90B 2026.03	56.2	53.9	-	-	-
Llama-3.2-90B 2026.04	56.2	53.9	42.6	57.3	61.7
Llama-3.2-90B 2026.04	56.2	53.9	42.6	57.3	61.7
Claude-3.5-Sonnet-(2024-06-22) 2026.03	55.3	53.6	-	-	-
Claude 3.5 Sonnet 2026.04	55.3	53.6	43.4	55	62.3
Claude-3.5-Sonnet (2024-06-22) 2026.04	55.3	53.6	43.4	55	62.3
InternVL3.5-14B 2026.03	54.5	-	56.8	65.4	41.4
Qwen2.5-VL-7B 2026.02	53.2	50.9	40.9	54.3	57.4
InternVL3.5-8B 2026.03	52.7	-	53.3	68	36.8
Qwen3-VL-8B-Instruct 2026.03	51.7	-	39	52	64.2
Qwen2.5-VL-72B-Instruct 2026.04	51.6	52.7	47.8	46.8	63.5
Qwen2.5-VL-72B 2026.04	51.16	52.73	48.07	46.73	63.41
MM-RLHF-Reward 2026.03	51	-	45	50.5	57.6
InternVL3-8B 2026.04	51	55.54	60.22	43.93	62.46
R1-Reward* 2026.04	50.8	46.6	43.1	57.5	39.1
LLaVA-Critic 2026.03	50.7	-	54.6	38.3	59.1
Eagle-2.5-8B 2026.02	50.2	49.7	41.4	48.6	59.3
MM-RLHF-Reward 2026.04	50.2	51	45	50.5	57.6
MM-RLHF-Reward 2026.02	50.15	-	-	-	-
MM-RLHF-Reward 2026.04	50.15	51.01	45.04	50.45	57.55
Qwen2.5-VL-7B-Instruct 2026.04	48	49.5	43.4	42	63
Cosmos-R1-7B 2026.02	44.8	44.8	33.1	41.3	59.9
Llama-3.2-11B 2026.04	42.9	42.8	33.3	38.4	56.6
Robobrain2.0-7B 2026.02	42.4	44.2	39.2	55.8	37.5
GPT-4o-mini 2026.04	41.5	44.8	41.7	34.5	58.2
LLaVA-Critic 2026.04	41.2	44	54.6	38.3	59.1
LLaVA-Critic 2026.04	41.2	44	54.6	38.3	59.1
NVLM-D-72B 2026.03	40.1	44.1	-	-	-
Qwen2-VL-72B 2026.02	39.5	-	-	-	-
Qwen2-VL-72B 2026.03	39.5	43	-	-	-
Qwen2-VL-72B 2026.04	39.5	43	38.1	32.8	58
Qwen2.5-VL-7B 2026.04	31.92	36.86	34.25	21.76	54.57
LLaVA-OneVision-7B 2026.04	29.6	36.5	32.2	20.1	57.1
Qwen2-VL-7B 2026.04	28.3	33.9	31.6	19.1	51.1
SliME 2026.02	19.04	-	-	-	-
SliME 2026.03	19.04	17.64	-	-	-

Showing 100 of 120 rows