Unverified50% confidenceBenchmarkExact time
In a six-question test on identifying tumors in brain scans, DeepSeek 4.1 Flash got all answers wrong; the evaluator noted that even the best models can only answer about one question correctly on this task.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/25/2026
First Seen
Valid until: 11/8/2026
Sources
Related Entities
Related Claims
UnverifiedIn the brain tumor identification task, the model's judgments of the tumor type in all six brain scan images were wrong.73% similarUnverified'Cherry-picked results' means researchers report only the results from among multiple experiments that support their hypothesis, which is a form of selective reporting.61% similarUnverifiedThis study is a preliminary pilot involving only 6 clinicians; the sample size is limited, and the generalizability of the conclusions awaits validation at a larger scale.59% similarUnverifiedIn imaging diagnosis, models may assess tumors based on spurious correlated features caused by dataset bias rather than genuine biomarkers.57% similarUnverifiedThe researchers selected four candidate datasets from HELM Safety that may target "harmful refusal", three of which are already saturated and cannot provide effective discriminating information.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1095714API
curl https://kongchang.com/api/v1/knowledge/claims/1095714MCP
get_claim(id=1095714)