Unverified50% confidenceEventExact time
A developer built a benchmark task in the Kaggle Benchmarking Challenge, using the IEEE-CIS fraud detection dataset to test Gemma 4 31B (google/gemma-4-31b) on its ability to judge whether a transaction is suspicious based solely on textualized graph structure information.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedA developer built a benchmark called Unlearning Security Audit in the Kaggle Benchmarking Challenge to test whether large language models can detect disguised unlearning code.77% similarUnverifiedAn offline audit of a Kaggle prompt injection evaluation built score auditing tools for three public Kaggle tasks to verify the true meaning of existing benchmark scores, rather than re-running models or training new systems.70% similarUnverified一位开发者在Kaggle Benchmarking Challenge中提交了一个新基准测试，衡量每个正确答案的成本而非单纯的模型价格69% similarUnverifiedDeveloper olgazju built a Kaggle benchmark called "Where Did My Data Go?" to test whether LLMs honestly admit they cannot determine the data's whereabouts when evidence is insufficient.68% similarUnverifiedSkills Store的审核机制包括检查是否存在提示词注入风险或数据泄露隐患68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1007678API
curl https://kongchang.com/api/v1/knowledge/claims/1007678MCP
get_claim(id=1007678)