M04 Evaluation Framework Base-line

Following development, students will rigorously evaluate the system's performance using a completed evaluation test set, running the evaluation pipeline against the batch interface, analyzing results, identifying errors, and proposing at least three specific improvement strategies.

• Completed evaluation test set (minimum 50–100 items, depending on format).
• Evaluation pipeline running against your batch interface.
• Baseline results + error analysis; at least three concrete improvement ideas.