Our new X account is live! Follow @wizwand_team for updates

Helpful Assistant Alignment on Helpful Assistant normalized rewards (test)

53Helpfulness Reward (r1)

MOPO-Lag

Updated 4d ago

Evaluation Results

Method	Links
MOPO-Lag 2025.12		53	51
MOPO-Lag 2025.12		52	53
MOPO-LB 2025.12		51	52
MOPO-LB 2025.12		49	48
MOPO-Lag 2025.12		49	47
MOPO-LB 2025.12		49	48
MOPO-Lag 2025.12		48	46
PARM 2025.12		48	47
PARM 2025.12		48	50
MOPO-LB 2025.12		45	43
MOPO-Lag 2025.12		45	47
PARM 2025.12		45	48
MOPO-Lag 2025.12		45	53
RiC 2025.12		45	47
RiC 2025.12		44	41
MOPO-LB 2025.12		44	45
MOPO-Lag 2025.12		43	41
RiC 2025.12		43	46
RiC 2025.12		43	41
RiC 2025.12		43	44
MOPO-Lag 2025.12		42	40
MOPO-LB 2025.12		42	50
MOPO-LB 2025.12		41	39
RiC 2025.12		41	47
MODPO 2025.12		41	39
PARM 2025.12		41	46
RiC 2025.12		40	33
PARM 2025.12		40	42
MOPO-LB 2025.12		40	39
MODPO 2025.12		40	42
MODPO 2025.12		40	40
MODPO 2025.12		39	37
PARM 2025.12		38	37
DPO on D_J 2025.12		38	41
MOPO-Lag 2025.12		37	36
RiC 2025.12		37	40
MODPO 2025.12		37	45
PARM 2025.12		36	39
DPO on D_J 2025.12		36	35
DPO on D_J 2025.12		36	39
MODPO 2025.12		35	33
MODPO 2025.12		35	33
PARM 2025.12		35	44
DPO on D_J 2025.12		35	40
DPO on D_J 2025.12		34	38
MOPO-LB 2025.12		34	35
PARM 2025.12		33	32
MODPO 2025.12		32	33
DPO on D_J 2025.12		32	30
DPO on D_J 2025.12		31	30
RiC 2025.12		31	27
MOPO-Lag 2025.12		29	26
PARM 2025.12		28	24
DPO on D_J 2025.12		28	26
MOPO-LB 2025.12		27	24
DPO on D_J 2025.12		27	25
MODPO 2025.12		25	24
RiC 2025.12		23	21
DPO on D_J 2025.12		22	19
MODPO 2025.12		21	20