CCTest · Blog
LIBERO-Industry Brings Factory Constraints to Robot Evaluation
Robotics & Physical AI
cctest.ai

LIBERO-Industry Brings Factory Constraints to Robot Evaluation

LIBERO-Industry extends the LIBERO benchmark with production-line conditions such as changing product geometry, moving conveyors, limited interaction windows, and tight placement tolerances. Tests show that generalist robot policies degrade steadily as these constraints are introduced and combined.

Read more
CCTest · Blog
BizBench Shows Why Long-Horizon Agents Must Be Judged on Process
Evaluation & Benchmarks
cctest.ai

BizBench Shows Why Long-Horizon Agents Must Be Judged on Process

BizBench brings real-world payment workflows into long-horizon agent evaluation, checking policy applicability, authorization timing, resource constraints, and intermediate compliance. Its results show that agents can reach a desired outcome while violating critical business conditions along the way.

Read more
CCTest · Blog
DuoDiscover Uses Fast Yardsticks to Break the Evaluation Bottleneck
Evaluation & Benchmarks
cctest.ai

DuoDiscover Uses Fast Yardsticks to Break the Evaluation Bottleneck

DuoDiscover is an autonomous search system for engineering optimization that lets coding agents discover both what to optimize and how to evaluate it cheaply. Its fast proxy evaluators are repeatedly checked against the real objective to prevent search from drifting toward misleading shortcuts.

Read more