{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreicbj3xc7abzgyupqicylbp4f6jef525bbb4g6s5ouej5awjdevwju",
"uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mnzsyisuri62"
},
"path": "/t/ai-safety-in-clinical-knowledge-graph/176454#post_4",
"publishedAt": "2026-06-11T17:07:08.000Z",
"site": "https://discuss.huggingface.co",
"textContent": "Yes, there is some similarity to routing / Mixture-of-Experts, but I think the distinction is important.\n\nIn MoE, the question is usually: which expert should process this input?\n\nIn DESi-style operator selection, the question is: which method is valid for this kind of evidence?\n\nFor example, a knowledge graph needs structural dependency checks, arithmetic needs deterministic calculation, current claims need retrieval, contaminated context needs segregation, and missing required inputs should trigger clarification instead of inference.\n\nSo I would frame it less as “which expert answers?” and more as “which verification procedure is allowed before an answer may exist?”",
"title": "AI safety in Clinical Knowledge graph"
}