specific method · filed under other
Removing Git-History Leaks from Benchmark Images
Remove leaked git history or future git tags from benchmark images to prevent reward hacking.
Also called removed to prevent possible reward hacking.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
we patched the base task images to remove the git history leak for all affected code_path benchmarks
usedevaluation onlyin LAGUNA M.1/XS.2Poolside
All images were updated to have future git tags removed to prevent possible reward hacking.
usedevaluation onlyin SWE-bench MultilingualPoolside
Filed alongside
Other methods under other.
Agentic WorkflowsAutomating Repetitive Infrastructure WorkClosed-Model Performance with Refusal SubstitutionCode-Driven AnimationEarly Release with Feedback-Driven IterationGenerated Artifact Validation with Linters and Runtime TestsJoint Optimization of Architecture, Cache Precision, and DeploymentPrompt Addendum Prohibiting Direct Online-Solution UsePrompt Modality OrderingRemoving Pattern-Matching-Based Anti-Cheat ChecksRequest Caching in the Browser ToolReward-Hacking Detection SystemScore-versus-Cost Efficiency ComparisonTraining–Inference ConsistencyUnique Run Identifiers