Unverified50% confidenceFactExact time
The researchers developed an evaluation framework comprising five complementary metrics: VCR, Model-Harness Fit, Cross-Model Fit, Interaction Residuals and Token Load.
1
Sources
50%
Confidence
Long-term
Relevance
10/26/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedModel-Harness Fit measures, for a given model, the extent to which a harness supports task achievement through valid completion; Cross-Model Fit measures a harness's overall applicability across all evaluated models.70% similarUnverifiedModelstamp 的五大核心能力包括：旁路清单、环境快照与相关包识别、反序列化前完整性校验、依赖漂移报告、可选HMAC认证69% similarUnverifiedThe research team evaluated over 35 compressed checkpoints across six open-weight models, covering the two mainstream method categories of quantization and pruning.69% similarUnverifiedThe research team designed five behavioral probes to test models' ability to satisfy hierarchical constraints.68% similarUnverifiedThe research team conducted extensive evaluations using multiple agent frameworks on both proprietary models and open-weight models; state-of-the-art models such as GLM-5.2 and Claude-Sonnet-4.6 achieved only slightly above passing performance.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1098741API
curl https://kongchang.com/api/v1/knowledge/claims/1098741MCP
get_claim(id=1098741)