{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreig2mlppdg4v2wkrej5xpdpsdjmi4cpszcg2bhl3jtymivufdtlf2q",
"uri": "at://did:plc:mxzzpugn7bprjjrszwkbez3u/app.bsky.feed.post/3mgne4zz33672"
},
"coverImage": {
"$type": "blob",
"ref": {
"$link": "bafkreieskizlxvrrkqgkdfhhtybfyshsde6oemuqjkqtzamdgcwfioa4fq"
},
"mimeType": "image/jpeg",
"size": 124701
},
"path": "/news/2026-03-ai-ability.html",
"publishedAt": "2026-03-09T11:20:03.000Z",
"site": "https://techxplore.com",
"tags": [
"Computer Sciences"
],
"textContent": "In high-stakes settings like medical diagnostics, users often want to know what led a computer vision model to make a certain prediction, so they can determine whether to trust its output. Concept bottleneck modeling is one method that enables artificial intelligence systems to explain their decision-making process. These methods force a deep-learning model to use a set of concepts, which can be understood by humans, to make a prediction. In new research, MIT computer scientists developed a method that coaxes the model to achieve better accuracy and clearer, more concise explanations.",
"title": "Improving AI models' ability to explain their predictions"
}