<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><description>PhD Student at Johns Hopkins University. Previously: Allen Institute for AI, Apple, Samaya AI. Research for #NLProc #IR</description><link>https://bsky.app/profile/orionweller.bsky.social</link><title>@orionweller.bsky.social - Orion Weller</title><item><link>https://bsky.app/profile/orionweller.bsky.social/post/3lj3n6ybtmc2v</link><description>Ever wonder how test-time compute would do in retrieval? 🤔&#xA;&#xA;introducing ✨rank1✨&#xA;&#xA;rank1 is distilled from R1 &amp; designed for reranking. &#xA;&#xA;rank1 is state-of-the-art at complex reranking tasks in reasoning, instruction-following, and general semantics (often 2x RankLlama 🤯)&#xA;&#xA;🧵</description><pubDate>26 Feb 2025 14:57 +0000</pubDate><guid isPermaLink="false">at://did:plc:vm5ic6kt3kljuttejxqf3dz3/app.bsky.feed.post/3lj3n6ybtmc2v</guid></item><item><link>https://bsky.app/profile/orionweller.bsky.social/post/3ldosvabhdk2r</link><description>Check out our new encoder model, ModernBERT! 🤖&#xA;&#xA;Super grateful to have been part of such an awesome team effort and very excited about the gains for retrieval/RAG! 🚀&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>19 Dec 2024 21:28 +0000</pubDate><guid isPermaLink="false">at://did:plc:vm5ic6kt3kljuttejxqf3dz3/app.bsky.feed.post/3ldosvabhdk2r</guid></item><item><link>https://bsky.app/profile/orionweller.bsky.social/post/3ldhbdlqgu42l</link><description>MASC is such a fun time! If your university is in the mid-Atlantic, please consider hosting!&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>16 Dec 2024 21:25 +0000</pubDate><guid isPermaLink="false">at://did:plc:vm5ic6kt3kljuttejxqf3dz3/app.bsky.feed.post/3ldhbdlqgu42l</guid></item><item><link>https://bsky.app/profile/orionweller.bsky.social/post/3lbngbpibvk2b</link><description>Creating a 🦋 starter pack for people working in IR/RAG: go.bsky.app/88ULgwY&#xA;&#xA;I can’t seem to find everyone though, help definitely appreciated to fill this out (DM or comment)!&#xA;&#xA;[contains quote post or other embedded content]</description><pubDate>23 Nov 2024 21:19 +0000</pubDate><guid isPermaLink="false">at://did:plc:vm5ic6kt3kljuttejxqf3dz3/app.bsky.feed.post/3lbngbpibvk2b</guid></item><item><link>https://bsky.app/profile/orionweller.bsky.social/post/3k7hdjhmxmh2n</link><description>Using LLMs for query or document expansion in retrieval (e.g. HyDE and Doc2Query) have scores going  📈&#xA;&#xA;But do these approaches work for all IR models and for different types of distribution shifts?  Turns out its actually more 📉 🚨&#xA;&#xA;📝 (arxiv soon): https://orionweller.github.io/assets/pdf/LLM_IR.pdf</description><pubDate>15 Sep 2023 18:57 +0000</pubDate><guid isPermaLink="false">at://did:plc:vm5ic6kt3kljuttejxqf3dz3/app.bsky.feed.post/3k7hdjhmxmh2n</guid></item></channel></rss>