Unverified50% confidenceBenchmarkExact time
SmolLM2-1.7B-Instruct achieved an overall score of 0.40 on EvidenceFlip, with conflict detection of 1.00 but pressure resistance of 0.00.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedQwen2.5-3B-Instruct achieved an overall score of 0.90 on EvidenceFlip, with an action accuracy of 0.90, conflict detection of 1.00, and pressure resistance of 1.00.78% similarUnverifiedQwen3.5-122B-A10B produced 8 refusals in 100 tests, with a KL divergence of 0.0856.59% similarUnverifiedThe PAANI detector checkpoint achieved a test mAP of 0.7388 at 0.5 IoU, and the rectangular ONNX export version achieved a validation mAP of 0.7367.58% similarUnverifiedOn a fixed revision of the repository, Praxa passed 1,027/1,027 unit tests and 89/89 Workerd tests, instrumented all 363 expected source files, and met four coverage lower bounds, though raw per-test records and independent reproduction are not yet available.58% similarUnverifiedThe fine-tuned 406M model achieved 36.33 ROUGE-1 on the test set, while the authors' own 1.2B system achieved 35.41 ROUGE-1, and the two cannot be statistically distinguished.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/999241API
curl https://kongchang.com/api/v1/knowledge/claims/999241MCP
get_claim(id=999241)