Unverified50% confidenceFactExact time
Scaling up model size or specifically building legal reasoning models (best-effort LoRA reproductions) cannot bridge the gap between the high correct-citation rate and the low decision sensitivity.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe paper points out that fidelity can only quantify whether an explanation can reproduce model outputs, but cannot guarantee that the explanation correctly reflects the underlying decision-making process.78% similarUnverifiedThe study explicitly states that neither increased model scale nor enhanced explicit reasoning capabilities provides resistance to this persuasion-based failure.76% similarUnverifiedThe moderation layer does not share weights with the main model and theoretically cannot be bypassed simultaneously by the same prompt.74% similarUnverifiedThe study finds that the marginal quality loss attributable to routing decisions themselves on this benchmark is not significant, and model bias is the more serious underlying problem.73% similarUnverifiedExisting evidence does not support the assumption that 'more capable models will better understand human intent and therefore be safer.'73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013278API
curl https://kongchang.com/api/v1/knowledge/claims/1013278MCP
get_claim(id=1013278)