List your company

Uncertainty Quantification on τ2-bench Retail

0.899AUROCSonnet-4.6
0.263560.428530.59350.75847Jun 24, 2026

Evaluation results

MethodLinks
Sonnet-4.6Training=false, Backbone=Qwen3.5-9B2026.060.899
Sonnet-4.6Training=false, Backbone=Qwen3-14B2026.060.864
Sonnet-4.6Training=false, Backbone=Gemma4-4B2026.060.852
Progress AdvantageTraining=false, Backbone=Gemma4-4B2026.060.69
WildReward-8BTraining=true, Backbone=Qwen3-14B2026.060.689
Progress AdvantageTraining=false, Backbone=Qwen3.5-9B2026.060.678
ThinkPRM-7BTraining=true, Backbone=Olmo3-7B2026.060.67
Progress AdvantageTraining=false, Backbone=Olmo3-7B2026.060.664
Sonnet-4.6Training=false, Backbone=Olmo3-7B2026.060.656
Progress AdvantageTraining=false, Backbone=Qwen3-14B2026.060.65
WildReward-8BTraining=true, Backbone=Gemma4-4B2026.060.643
ThinkPRM-14BTraining=true, Backbone=Qwen3-14B2026.060.637
ThinkPRM-14BTraining=true, Backbone=Qwen3.5-9B2026.060.61
Self-CertaintyTraining=false, Backbone=Qwen3-14B2026.060.608
DeepConf TailTraining=false, Backbone=Olmo3-7B2026.060.608
WildReward-8BTraining=true, Backbone=Olmo3-7B2026.060.584
DeepConf B10Training=false, Backbone=Qwen3-14B2026.060.582
ThinkPRM-14BTraining=true, Backbone=Gemma4-4B2026.060.573
ThinkPRM-7BTraining=true, Backbone=Qwen3.5-9B2026.060.551
ThinkPRM-14BTraining=true, Backbone=Olmo3-7B2026.060.544
ThinkPRM-7BTraining=true, Backbone=Qwen3-14B2026.060.543
DeepConf B10Training=false, Backbone=Qwen3.5-9B2026.060.496
ThinkPRM-7BTraining=true, Backbone=Gemma4-4B2026.060.469
WildReward-8BTraining=true, Backbone=Qwen3.5-9B2026.060.468
DeepConf B10Training=false, Backbone=Gemma4-4B2026.060.416
Self-CertaintyTraining=false, Backbone=Gemma4-4B2026.060.397
Self-CertaintyTraining=false, Backbone=Olmo3-7B2026.060.392
DeepConf TailTraining=false, Backbone=Gemma4-4B2026.060.382
DeepConf TailTraining=false, Backbone=Qwen3-14B2026.060.38
Self-CertaintyTraining=false, Backbone=Qwen3.5-9B2026.060.366
DeepConf TailTraining=false, Backbone=Qwen3.5-9B2026.060.344
DeepConf B10Training=false, Backbone=Olmo3-7B2026.060.288