<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: apwheele</title><link>https://news.ycombinator.com/user?id=apwheele</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Thu, 10 Sep 2026 09:25:42 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=apwheele" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[Springer AI reference checks are not working]]></title><description><![CDATA[
<p>Article URL: <a href="https://veruscite.com/blog/springer-ai-checks-are-not-working">https://veruscite.com/blog/springer-ai-checks-are-not-working</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49629120">https://news.ycombinator.com/item?id=49629120</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 09 Sep 2026 16:28:04 +0000</pubDate><link>https://veruscite.com/blog/springer-ai-checks-are-not-working</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49629120</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49629120</guid></item><item><title><![CDATA[New comment by apwheele in "Flock Wants a Closely Surveilled World with No Exit"]]></title><description><![CDATA[
<p>They are relevant and courts have considered the whole of a persons movements standard in terms of whether querying ALPR data constitutes a search, <a href="https://andrewpwheeler.com/2026/08/12/license-plate-reader-searches-should-require-a-warrant/" rel="nofollow">https://andrewpwheeler.com/2026/08/12/license-plate-reader-s...</a><p>Under current case law (Carpenter and recently affirmed in Chatrie) it will definitely be a search, IMO it is just when the sensors become dense enough according to the court (absent states do not make regulation themselves to require a warrant for historical data).</p>
]]></description><pubDate>Wed, 09 Sep 2026 12:41:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=49625601</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49625601</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49625601</guid></item><item><title><![CDATA[New comment by apwheele in "Hallucinated Citations and Accuracy of Papers"]]></title><description><![CDATA[
<p>I sent it to the moderation email this morning. Will update if I hear back.</p>
]]></description><pubDate>Sat, 05 Sep 2026 17:29:32 +0000</pubDate><link>https://news.ycombinator.com/item?id=49578699</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49578699</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49578699</guid></item><item><title><![CDATA[Hallucinated Citations and Accuracy of Papers]]></title><description><![CDATA[
<p>Article URL: <a href="https://veruscite.com/blog/hallucinated-citations-and-accuracy-of-papers">https://veruscite.com/blog/hallucinated-citations-and-accuracy-of-papers</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49578452">https://news.ycombinator.com/item?id=49578452</a></p>
<p>Points: 2</p>
<p># Comments: 3</p>
]]></description><pubDate>Sat, 05 Sep 2026 17:04:28 +0000</pubDate><link>https://veruscite.com/blog/hallucinated-citations-and-accuracy-of-papers</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49578452</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49578452</guid></item><item><title><![CDATA[The high cost of free data: Why are cities paying so much for open data portals?]]></title><description><![CDATA[
<p>Article URL: <a href="https://grandschema.substack.com/p/the-high-cost-of-free-data">https://grandschema.substack.com/p/the-high-cost-of-free-data</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49539346">https://news.ycombinator.com/item?id=49539346</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Wed, 02 Sep 2026 17:14:21 +0000</pubDate><link>https://grandschema.substack.com/p/the-high-cost-of-free-data</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49539346</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49539346</guid></item><item><title><![CDATA[Some Cities Reward Careful Drivers with Fewer Red Lights]]></title><description><![CDATA[
<p>Article URL: <a href="https://apnews.com/article/rest-red-reward-drivers-portland-albuquerque-bcec5e0fe91dcaaaeb8f60e866ff7a37">https://apnews.com/article/rest-red-reward-drivers-portland-albuquerque-bcec5e0fe91dcaaaeb8f60e866ff7a37</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49520682">https://news.ycombinator.com/item?id=49520682</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 01 Sep 2026 11:48:53 +0000</pubDate><link>https://apnews.com/article/rest-red-reward-drivers-portland-albuquerque-bcec5e0fe91dcaaaeb8f60e866ff7a37</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49520682</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49520682</guid></item><item><title><![CDATA[New comment by apwheele in "Your AGENTS.md file doesn't do anything"]]></title><description><![CDATA[
<p>This is what the majority of mine look like as well. Just simple instructions for things I need to do repeatedly.<p>I have not even really needed a formal memory system. If I see an error happen more than once, I just say "hey add a note on this to agents.md". Tends to be verbose but overall works quite well for the projects I am doing.</p>
]]></description><pubDate>Fri, 28 Aug 2026 12:16:25 +0000</pubDate><link>https://news.ycombinator.com/item?id=49477414</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49477414</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49477414</guid></item><item><title><![CDATA[New comment by apwheele in "Gemini Omni 1.1 Flash"]]></title><description><![CDATA[
<p>Is there a good example someone can show of naturally generating real videos using extending and interpolation from the last frames? (For any of the video generators).<p>I mean I have not tinkered that much, but trying to even get a video to 30 seconds (I just want a cartoon AI avatar to narrate tutorials) has been incredibly difficult. They drift so easily.<p>Many AI videos you can tell just stitch short clips together. I just want a continuous scene for like 30 seconds to 2 minutes.</p>
]]></description><pubDate>Fri, 28 Aug 2026 12:13:14 +0000</pubDate><link>https://news.ycombinator.com/item?id=49477393</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49477393</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49477393</guid></item><item><title><![CDATA[Why I don't use Google Scholar in VerusCite]]></title><description><![CDATA[
<p>Article URL: <a href="https://veruscite.com/blog/why-i-dont-use-google-scholar">https://veruscite.com/blog/why-i-dont-use-google-scholar</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49467357">https://news.ycombinator.com/item?id=49467357</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Thu, 27 Aug 2026 16:24:20 +0000</pubDate><link>https://veruscite.com/blog/why-i-dont-use-google-scholar</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49467357</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49467357</guid></item><item><title><![CDATA[Hall of Hallucinated Citations in Published Papers]]></title><description><![CDATA[
<p>Article URL: <a href="https://veruscite.com/hallucinations">https://veruscite.com/hallucinations</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49436254">https://news.ycombinator.com/item?id=49436254</a></p>
<p>Points: 3</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 25 Aug 2026 15:57:41 +0000</pubDate><link>https://veruscite.com/hallucinations</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49436254</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49436254</guid></item><item><title><![CDATA[A Google DeepMind paper has multiple hallucinated references]]></title><description><![CDATA[
<p>Article URL: <a href="https://veruscite-data.com/share/-7KyAgTQ6aqub5ilvQ_6_SVg2wkbjhp95pLeG3QFa-o">https://veruscite-data.com/share/-7KyAgTQ6aqub5ilvQ_6_SVg2wkbjhp95pLeG3QFa-o</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49409192">https://news.ycombinator.com/item?id=49409192</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 23 Aug 2026 14:37:05 +0000</pubDate><link>https://veruscite-data.com/share/-7KyAgTQ6aqub5ilvQ_6_SVg2wkbjhp95pLeG3QFa-o</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49409192</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49409192</guid></item><item><title><![CDATA[Notes on Crime Trends in the US]]></title><description><![CDATA[
<p>Article URL: <a href="https://crimede-coder.com/blogposts/2026/CrimeTrends">https://crimede-coder.com/blogposts/2026/CrimeTrends</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49344450">https://news.ycombinator.com/item?id=49344450</a></p>
<p>Points: 4</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 18 Aug 2026 12:04:05 +0000</pubDate><link>https://crimede-coder.com/blogposts/2026/CrimeTrends</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49344450</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49344450</guid></item><item><title><![CDATA[New comment by apwheele in "Don't classify, hallucinate!"]]></title><description><![CDATA[
<p>This is another riff on not embedding a full document, but doing a summarization of the document and embedding the summary for RAG. Nice usecase for high cardinality data!</p>
]]></description><pubDate>Fri, 14 Aug 2026 13:29:34 +0000</pubDate><link>https://news.ycombinator.com/item?id=49298418</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49298418</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49298418</guid></item><item><title><![CDATA[New comment by apwheele in "Flock updates privacy, accountability, security, and transparency safeguards"]]></title><description><![CDATA[
<p>I did not know about this when I wrote <a href="https://andrewpwheeler.com/2026/08/12/license-plate-reader-searches-should-require-a-warrant/" rel="nofollow">https://andrewpwheeler.com/2026/08/12/license-plate-reader-s...</a> the day prior, but one of the things I did state was deleting data (except in the extreme case, like New Hampshire's 3 minute retention), will not prevent abuse. People just do illegal queries over and over again.<p>Part of the reason I wrote that post is it is better to retain the data indefinitely and have a warrant standard than continue to reduce the time period in which the data is retained.</p>
]]></description><pubDate>Fri, 14 Aug 2026 12:10:11 +0000</pubDate><link>https://news.ycombinator.com/item?id=49297693</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49297693</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49297693</guid></item><item><title><![CDATA[New comment by apwheele in "License plate reader searches should require a warrant"]]></title><description><![CDATA[
<p>This is already the current status quo on paper. I bet the majority of the cases IJ identified of mis-use the PDs already had this policy. (That is the same standard when you do a background criminal history check in New York, need to put in a case number, FYI.)<p>I did not want to go into too many technical details on the post -- I think you could do an AI audit at the point in time of the search query to prevent people putting in junk cases. That said I believe there will always need to be third party audits at a minimum.<p>Part of the issue is policies on paper are not effectively enforced. So people saying "just have policy X to prevent abuse" is a non-answer.<p>I wrote the post mainly because people arguing for limited data retention I think is bad for both sides -- it neither protects civil liberties and simultaneously makes it harder for long term criminal investigations.</p>
]]></description><pubDate>Wed, 12 Aug 2026 18:31:52 +0000</pubDate><link>https://news.ycombinator.com/item?id=49276720</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49276720</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49276720</guid></item><item><title><![CDATA[License plate reader searches should require a warrant]]></title><description><![CDATA[
<p>Article URL: <a href="https://andrewpwheeler.com/2026/08/12/license-plate-reader-searches-should-require-a-warrant/">https://andrewpwheeler.com/2026/08/12/license-plate-reader-searches-should-require-a-warrant/</a></p>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49273165">https://news.ycombinator.com/item?id=49273165</a></p>
<p>Points: 644</p>
<p># Comments: 397</p>
]]></description><pubDate>Wed, 12 Aug 2026 14:43:39 +0000</pubDate><link>https://andrewpwheeler.com/2026/08/12/license-plate-reader-searches-should-require-a-warrant/</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49273165</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49273165</guid></item><item><title><![CDATA[Show HN: VerusCite, a tool to check for hallucinations in academic articles]]></title><description><![CDATA[
<p>Hello HackerNews --<p>I have created an application that verifies citations. Main motivation is to make checking for hallucinated citations in academic papers simpler/faster.<p>Can sign up for the tool at <<a href="https://veruscite-data.com/" rel="nofollow">https://veruscite-data.com/</a>>, and can see the benchmark for metrics and some description of the methodology at <<a href="https://github.com/apwheele/veruscite-data" rel="nofollow">https://github.com/apwheele/veruscite-data</a>>.<p>If you want to check out the approximate UI without signing up (at least the output), can see some examples:<p><a href="https://veruscite-data.com/share/WWcj0furPRsDa8wlYKQSER1zQYp" rel="nofollow">https://veruscite-data.com/share/WWcj0furPRsDa8wlYKQSER1zQYp</a>...<p><a href="https://veruscite-data.com/share/xnHTxfP6bp7p3ZKdQegWuNvyQtX" rel="nofollow">https://veruscite-data.com/share/xnHTxfP6bp7p3ZKdQegWuNvyQtX</a>...<p><a href="https://veruscite-data.com/share/3Pu3dwCfc-ixPgtHutVFXgMU7FY" rel="nofollow">https://veruscite-data.com/share/3Pu3dwCfc-ixPgtHutVFXgMU7FY</a>...<p>There are many minor errors in citation lists (maybe on the order of 5%). So even if you were not worried about hallucinations, it is useful for editors. If you sign up I have attempted to make the UI as easy as possible for a human to review the output and make changes (there are definitely errors, see the github for estimates of error rates).<p>So let me know what you think HN!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49234321">https://news.ycombinator.com/item?id=49234321</a></p>
<p>Points: 1</p>
<p># Comments: 0</p>
]]></description><pubDate>Sun, 09 Aug 2026 18:45:45 +0000</pubDate><link>https://veruscite-data.com/login?next=%2Fdashboard</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49234321</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49234321</guid></item><item><title><![CDATA[New comment by apwheele in "New Orleans is testing Carbyne’s AI-powered Emergency Call Triage software"]]></title><description><![CDATA[
<p>I cannot read this article, but this may be partly confused what this function by Carbyne does. If there is a flood of critical systems, it just plays an auto message "If this is about X, thank you and we know about it" if you called from the same area.<p>I don't remember from prior articles if at that point you just stay on the line, or if you need to press a button. It is not going through an AI voice call tree.</p>
]]></description><pubDate>Fri, 07 Aug 2026 12:04:15 +0000</pubDate><link>https://news.ycombinator.com/item?id=49209124</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49209124</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49209124</guid></item><item><title><![CDATA[Show HN: VerusCite, a tool to check for hallucinations in academic articles]]></title><description><![CDATA[
<p>Hello HackerNews --<p>I have created an application that verifies citations. Main motivation is to make checking for hallucinated citations in academic papers simpler/faster.<p>Can sign up for the tool at <<a href="https://veruscite-data.com/" rel="nofollow">https://veruscite-data.com/</a>>, and can see the benchmark for metrics and some description of the methodology at <<a href="https://github.com/apwheele/veruscite-data" rel="nofollow">https://github.com/apwheele/veruscite-data</a>>.<p>If you want to check out the approximate UI without signing up (at least the output), can see some examples:<p><a href="https://veruscite-data.com/share/WWcj0furPRsDa8wlYKQSER1zQYpZBi2t--7jH75eKW8" rel="nofollow">https://veruscite-data.com/share/WWcj0furPRsDa8wlYKQSER1zQYp...</a><p><a href="https://veruscite-data.com/share/xnHTxfP6bp7p3ZKdQegWuNvyQtXgs6INRM6oUyv0LPM" rel="nofollow">https://veruscite-data.com/share/xnHTxfP6bp7p3ZKdQegWuNvyQtX...</a><p><a href="https://veruscite-data.com/share/3Pu3dwCfc-ixPgtHutVFXgMU7FYw9qApSdDpMvscLZ8" rel="nofollow">https://veruscite-data.com/share/3Pu3dwCfc-ixPgtHutVFXgMU7FY...</a><p>There are many minor errors in citation lists (maybe on the order of 5%). So even if you were not worried about hallucinations, it is useful for editors. If you sign up I have attempted to make the UI as easy as possible for a human to review the output and make changes (there are definitely errors, see the github for estimates of error rates).<p>So let me know what you think HN!</p>
<hr>
<p>Comments URL: <a href="https://news.ycombinator.com/item?id=49172077">https://news.ycombinator.com/item?id=49172077</a></p>
<p>Points: 2</p>
<p># Comments: 0</p>
]]></description><pubDate>Tue, 04 Aug 2026 17:30:49 +0000</pubDate><link>https://veruscite-data.com/</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49172077</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49172077</guid></item><item><title><![CDATA[New comment by apwheele in "Why Large Language Models Fail at Tabular Prediction"]]></title><description><![CDATA[
<p>My money is on they used agentic AI coding tools to help set up all the metrics/experiments with traditional models.</p>
]]></description><pubDate>Tue, 04 Aug 2026 16:33:33 +0000</pubDate><link>https://news.ycombinator.com/item?id=49171241</link><dc:creator>apwheele</dc:creator><comments>https://news.ycombinator.com/item?id=49171241</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49171241</guid></item></channel></rss>