<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><description>machine learning for health at microsoft research, based in cambridge UK 🌻 she/her</description><link>https://bsky.app/profile/hylandsl.bsky.social</link><title>@hylandsl.bsky.social - Stephanie Hyland</title><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3mk7uca6oc224</link><description>The #ICLR2026 fomo is already intense</description><pubDate>24 Apr 2026 06:06 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3mk7uca6oc224</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3mchbid7jak27</link><description>the greatest joy of being a computational scientist is having the computer work for you while you do something else</description><pubDate>15 Jan 2026 09:29 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3mchbid7jak27</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3m7rmvueqfs2q</link><description>“Interpretability plays a special role in machine learning because instead of focusing on making the AI smarter, we focus on improving human insight. I think this is the most important category of interpretability research, and we do not do enough of it.”&#xA;&#xA;😎😎😎&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>12 Dec 2025 08:04 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3m7rmvueqfs2q</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3m76eqhqepc2n</link><description>remember to always include a QR code on your poster. spotted at #eurips</description><pubDate>04 Dec 2025 16:18 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3m76eqhqepc2n</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3m6cs6uvguk2h</link><description>found a file from PhD days with the FORTY-EIGHT ways &#34;ACE inhibitor&#34; was encoded in the EHR system we were working wth</description><pubDate>23 Nov 2025 17:04 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3m6cs6uvguk2h</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3m5r7ogsqyk2f</link><description>finally got around to booking my travel for #EurIPS2025! Looking forward to connecting with the European ML scene in Copenhagen</description><pubDate>16 Nov 2025 17:17 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3m5r7ogsqyk2f</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lzewqx76cs2z</link><description>uv is so good&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>21 Sep 2025 22:25 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lzewqx76cs2z</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lwejcaz6oc2p</link><description>what is the purpose of VQA datasets where text-only models do better than random?</description><pubDate>14 Aug 2025 14:08 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lwejcaz6oc2p</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lwbwuyhkyk2c</link><description>lads can we stop</description><pubDate>13 Aug 2025 13:34 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lwbwuyhkyk2c</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lv6ztnwugc26</link><description>it&#39;s frustrating how inefficient review assignments are: we target a minimum number of completed reviews per paper but in accounting for inevitable no-shows, some people end up doing technically unnecessary (if still beneficial) reviews</description><pubDate>30 Jul 2025 16:23 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lv6ztnwugc26</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lua55lreg22g</link><description>New work from my team! arxiv.org/abs/2507.12950&#xA;Intersecting mechanistic interpretability and health AI 😎&#xA;&#xA;We trained and interpreted sparse autoencoders on MAIRA-2, our radiology MLLM. We found a range of human-interpretable radiology reporting concepts, but also many uninterpretable SAE features.&#xA;https://arxiv.org/abs/2507.12950</description><pubDate>18 Jul 2025 09:30 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lua55lreg22g</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lslfgmfpsk2v</link><description>This is cool arxiv.org/abs/2505.18235.&#xA;Linear representation hypothesis discourse needs more differential geometry i m o&#xA;https://arxiv.org/abs/2505.18235</description><pubDate>27 Jun 2025 10:07 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lslfgmfpsk2v</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lrvftaeeak2m</link><description>Limited time offer: I get an email every time someone fills in this form, so act now to add to the chaos of my inbox before I figure out how to turn this off&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>18 Jun 2025 16:16 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lrvftaeeak2m</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lrncphbbps2v</link><description>A strange thing about living near Duxford is having WW2-era planes flying overhead on a regular basis.</description><pubDate>15 Jun 2025 10:59 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lrncphbbps2v</guid></item><item><link>https://bsky.app/profile/hylandsl.bsky.social/post/3lqcelezplk2c</link><description>NeurIPS 2025 call for ethics reviewers https://neurips.cc/Conferences/2025/CallForEthicsReviewers&#xA;#neurips2025</description><pubDate>29 May 2025 09:08 +0000</pubDate><guid isPermaLink="false">at://did:plc:jdtvwiusvyitistp225yzqh2/app.bsky.feed.post/3lqcelezplk2c</guid></item></channel></rss>