List your company
ResearchFormerly WizWand

Planning on Countdown

87.5AccuracyBGPO
4.61226.13147.6569.169Oct 13, 2025Nov 23, 2025Jan 3, 2026Feb 13, 2026Mar 26, 2026May 6, 2026Jun 16, 2026
Updated 4mo ago

Evaluation results

MethodLinks
BGPOBase Model=LLaDA-8B-Instruct, RL Algorithm=BGPO2025.1087.5
VRPO-OLBase Model=LLaDA-8B-Instruct, RL Algorithm=VRPO-OL, Sampling size (nt)=42025.1084.8
BGPO (off-policy)Base Model=LLaDA-8B-Instruct, RL Algorithm=BGPO, Mode=off-policy2025.1084.8
ESPOBackbone=LLaDA-8B, Seq Len=2562025.1282
ESPOBackbone=LLaDA-8B, Seq Len=1282025.1281.6
ESPOBackbone=LLaDA-8B, Seq Len=Avg.2025.1281
ESPOBackbone=LLaDA-8B, Seq Len=5122025.1279.3
RSPOGeneration length=2562026.0578.83
RSPOGeneration length=5122026.0573.83
ESPOBackbone=Dream-7B, Seq Len=1282025.1268.8
ESPOBackbone=Dream-7B, Seq Len=2562025.1266.8
ESPOBackbone=Dream-7B, Seq Len=Avg.2025.1266.8
ESPOBackbone=Dream-7B, Seq Len=5122025.1264.8
SAPOGeneration length=5122026.0556.3
diffu-GRPOBase Model=LLaDA-8B-Instruct, RL Algorithm=diffu-GRPO2025.1053.1
TraceRLGeneration length=5122026.0552.6
SAPOGeneration length=2562026.0552
wd1Backbone=LLaDA-8B, Seq Len=2562025.1251.2
wd1Generation length=2562026.0551.2
TraceRLGeneration length=2562026.0550.4
wd1Backbone=LLaDA-8B, Seq Len=Avg.2025.1248.3
wd1Backbone=LLaDA-8B, Seq Len=1282025.1247.7
wd1Backbone=LLaDA-8B, Seq Len=5122025.1246.1
wd1Generation length=5122026.0546.1
wd12025.1046.1
Info-Gain SamplerK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0245.2
Info-Gain SamplerK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0244.1
PC-SamplerK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0242.4
PC-Sampler w/ E-BONModel=LLaDA-Instruct-8B, Decoding Strategy=PC-Sampler + E-BON2025.1242.2
wd1Backbone=Dream-7B, Seq Len=5122025.1242.2
PC-SamplerK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0241.8
PC-Sampler w/ E-SMCModel=LLaDA-Instruct-8B, Decoding Strategy=PC-Sampler + E-SMC2025.1240.4
PC-Sampler w/ E-BONModel=LLaDA-1.5-8B, Decoding Strategy=PC-Sampler + E-BON2025.1240.2
PC-Sampler w/ E-SMCModel=LLaDA-1.5-8B, Decoding Strategy=PC-Sampler + E-SMC2025.1239.3
ConfidenceK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0239.2
wd1Backbone=Dream-7B, Seq Len=2562025.1237.9
d-OPSDSeq Len=1282026.0637.9
diffu-GRPOBackbone=Dream-7B, Seq Len=5122025.1237.5
diffu-GRPOBackbone=LLaDA-8B, Seq Len=5122025.1237.1
d1Generation length=5122026.0537.1
d3LLMSeq Len=1282026.0636.7
PC-SamplerModel=LLaDA-Instruct-8B, Decoding Strategy=PC-Sampler2025.1236.3
wd1Backbone=Dream-7B, Seq Len=Avg.2025.1236.3
ConfidenceK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0236.2
KLASSK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0235.4
MarginK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0235.2
PC-SamplerModel=LLaDA-1.5-8B, Decoding Strategy=PC-Sampler2025.1235
EntropyModel=LLaDA-1.5-8B, Decoding Strategy=Entropy2025.1234.7
EB-SamplerModel=LLaDA-1.5-8B, Decoding Strategy=EB-Sampler2025.1234.6
KLASSK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0234.3
P2Model=LLaDA-Instruct-8B, Decoding Strategy=P22025.1234.2
EB-SamplerModel=LLaDA-Instruct-8B, Decoding Strategy=EB-Sampler2025.1234.1
ConfidenceModel=LLaDA-Instruct-8B, Decoding Strategy=Confidence2025.1234
MarginModel=LLaDA-Instruct-8B, Decoding Strategy=Margin2025.1233.9
diffu-GRPOBackbone=LLaDA-8B, Seq Len=Avg.2025.1233.9
EntropyModel=LLaDA-Instruct-8B, Decoding Strategy=Entropy2025.1233.8
ConfidenceModel=LLaDA-1.5-8B, Decoding Strategy=Confidence2025.1233.8
P2Model=LLaDA-1.5-8B, Decoding Strategy=P22025.1233.6
EntropyK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0233.6
diffu-GRPOBackbone=LLaDA-8B, Seq Len=1282025.1233.2
diffu-GRPOSeq Len=1282026.0633.2
Fast-dLLMModel=LLaDA-1.5-8B, Decoding Strategy=Fast-dLLM2025.1232.9
Semi-ARModel=LLaDA-Instruct-8B, Decoding Strategy=Semi-AR2025.1232.6
Semi-ARModel=LLaDA-1.5-8B, Decoding Strategy=Semi-AR2025.1232.4
d-OPSDSeq Len=2562026.0632.3
d1-LLaDA2025.1032
MarginModel=LLaDA-1.5-8B, Decoding Strategy=Margin2025.1231.8
diffu-GRPOBackbone=LLaDA-8B, Seq Len=2562025.1231.3
d1Generation length=2562026.0531.3
diffu-GRPOSeq Len=2562026.0631.3
diffu-GRPOBackbone=Dream-7B, Seq Len=Avg.2025.1230.8
wd1Backbone=Dream-7B, Seq Len=1282025.1228.9
MarginK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0228.9
diffu-GRPOBackbone=Dream-7B, Seq Len=2562025.1227.7
diffu-GRPOBackbone=Dream-7B, Seq Len=1282025.1227.3
EntropyK=2, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.0223.2
VRPOGeneration length=2562026.0522.3
VRPOSeq Len=1282026.0621.9
VRPOSeq Len=2562026.0621.1
LLaDABackbone=LLaDA-8B, Seq Len=1282025.1220.7
LLaDA-8B-InstructSeq Len=1282026.0620.3
SFT VariantSeq Len=1282026.0620.3
LLaDABackbone=LLaDA-8B, Seq Len=2562025.1219.5
LLaDA-8B-InstructGeneration length=2562026.0519.5
LLaDA-8B-InstructBase Model=LLaDA-8B-Instruct2025.1019.5
LLaDA-8B-InstructSeq Len=2562026.0619.1
LLaDABackbone=LLaDA-8B, Seq Len=Avg.2025.1218.7
VRPOGeneration length=5122026.0518
DreamBackbone=Dream-7B, Seq Len=5122025.1217.4
LLaDABackbone=LLaDA-8B, Seq Len=5122025.1216
LLaDA-8B-InstructGeneration length=5122026.0516
UniformModel=LLaDA-1.5-8B, Decoding Strategy=Uniform2025.1215.8
SFT VariantSeq Len=2562026.0614.5
UniformModel=LLaDA-Instruct-8B, Decoding Strategy=Uniform2025.1214.4
d3LLMSeq Len=2562026.0612.5
Fast-dLLMModel=LLaDA-Instruct-8B, Decoding Strategy=Fast-dLLM2025.1211.4
DreamBackbone=Dream-7B, Seq Len=Avg.2025.1211.2
UniformK=1, Backbone=Dream-7B, Decoding Strategy=Global, Attention Type=full-attention2026.028.8
DreamBackbone=Dream-7B, Seq Len=1282025.128.5
DreamBackbone=Dream-7B, Seq Len=2562025.127.8