Dynamic Task Pool Sizing in DKD for CodeT5 on MBXP Benchmark Performance

This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How does varying the size of the task pool in DKD impact CodeT5's performance on the MBXP benchmark in terms of accuracy and training throughput compared to fixed-size task pools in continual. 8 claims were extracted from source literature; 8 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 8.3/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: How does varying the size of the task pool in DKD impact CodeT5's performance on the MBXP benchmark in terms of accuracy and training throughput compared to fixed-size task pools in continual learning scenarios?

Autonomous literature synthesis. Automated review score: 8.3/10. Full text and citation available at Assignee Research.