uchihamadara1816's picture
Upload 172 files
d02bacd verified

AutoDataLab++ Method Context Summary

Method Task RAG Model route Fallback Auto-finish Policy reward Terminal Consulted
sft expert_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2827 0.8827 analyst, finance, strategy, hr
sft risk_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2839 0.8839 analyst, finance, strategy, hr
sft crisis_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2805 0.8805 analyst, finance, strategy, hr
sft expert_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2925 0.8925 analyst, finance, strategy, hr
sft risk_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2948 0.8948 analyst, finance, strategy, hr
sft crisis_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2914 0.8914 analyst, finance, strategy, hr
dpo expert_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8827 analyst, finance, strategy, hr
dpo risk_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8839 analyst, finance, strategy, hr
dpo crisis_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8805 analyst, finance, strategy, hr
dpo expert_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8925 analyst, finance, strategy, hr
dpo risk_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8948 analyst, finance, strategy, hr
dpo crisis_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:strategy -> consult:strategy -> consult:strategy consult:hr -> summarize -> submit - -0.26 0.8914 analyst, finance, strategy, hr
sft_dpo expert_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2827 0.8827 analyst, finance, strategy, hr
sft_dpo risk_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2839 0.8839 analyst, finance, strategy, hr
sft_dpo crisis_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2805 0.8805 analyst, finance, strategy, hr
sft_dpo expert_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2925 0.8925 analyst, finance, strategy, hr
sft_dpo risk_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2948 0.8948 analyst, finance, strategy, hr
sft_dpo crisis_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2914 0.8914 analyst, finance, strategy, hr
grpo_rlvr expert_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2827 0.8827 analyst, finance, strategy, hr
grpo_rlvr risk_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2839 0.8839 analyst, finance, strategy, hr
grpo_rlvr crisis_brief False consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2805 0.8805 analyst, finance, strategy, hr
grpo_rlvr expert_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2925 0.8925 analyst, finance, strategy, hr
grpo_rlvr risk_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2948 0.8948 analyst, finance, strategy, hr
grpo_rlvr crisis_brief True consult:analyst -> consult:finance -> consult:strategy -> consult:hr -> summarize -> submit - - 1.2914 0.8914 analyst, finance, strategy, hr