DiffCodeGen clusters code candidates by behavioral similarity from fuzzing-synthesized inputs and selects the largest cluster's medoid, matching or exceeding prior test-time scaling methods with far less token and time cost.
Holistic Evaluation of Language Models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Code Generation by Differential Test Time Scaling
DiffCodeGen clusters code candidates by behavioral similarity from fuzzing-synthesized inputs and selects the largest cluster's medoid, matching or exceeding prior test-time scaling methods with far less token and time cost.