SWE-Cycle benchmark shows sharp drops in code agent success rates from isolated tasks to full autonomous issue resolution, highlighting cross-phase dependency issues.
Both tests pass: test_filter_book_limit_constant_exists and test_solr_opts_has_boolean_clauses_limit
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
-
SWE-Cycle: Benchmarking Code Agents across the Complete Issue Resolution Cycle
SWE-Cycle benchmark shows sharp drops in code agent success rates from isolated tasks to full autonomous issue resolution, highlighting cross-phase dependency issues.