This repository provides merged full-parameter judger models for deception evaluation in agent behaviors, released as part of SPADE-Bench.
-
plan_judger_lora_32B/
A judger model specialized in classifying the stance of declared plans.
-
action_judger_lora_32B/
A judger model specialized in classifying the stance of executed actions.
1vllm serve <model_path> \
2 --served-model-name <name> \
3 --enable-auto-tool-choice \
4 --tool-call-parser hermes \
5 --port <port>
6
