Area woman continuously working in recsys more and more convinced that most ML problems are recsys problems
September 7, 2026 at 8:51 PM
Area woman continuously working in recsys more and more convinced that most ML problems are recsys problems
Engagement KPIs have been responsible for infinite scroll UI traps, RecSys radicalization spirals, public shame brigades, and basically every modern ailment novel to the past decade. At some point, you have to recognize that the problem isn't any one technology, it's the metric.
The company essentially turned a dial that made ChatGPT more appealing and made people use it more, but sent some of them into delusional spirals.
OpenAI has since made the chatbot safer, but that comes with a tradeoff: less usage.
OpenAI has since made the chatbot safer, but that comes with a tradeoff: less usage.
November 23, 2025 at 8:12 PM
Engagement KPIs have been responsible for infinite scroll UI traps, RecSys radicalization spirals, public shame brigades, and basically every modern ailment novel to the past decade. At some point, you have to recognize that the problem isn't any one technology, it's the metric.
What industrial recsys papers have you enjoyed or found useful in the past year or two? Sharing my list:
# 1. Integrating LLMs into recsys
1.1. LLM-augmented recommenders
• Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations
- arxiv.org/abs/2306.08121
# 1. Integrating LLMs into recsys
1.1. LLM-augmented recommenders
• Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations
- arxiv.org/abs/2306.08121
January 31, 2025 at 3:23 AM
What industrial recsys papers have you enjoyed or found useful in the past year or two? Sharing my list:
# 1. Integrating LLMs into recsys
1.1. LLM-augmented recommenders
• Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations
- arxiv.org/abs/2306.08121
# 1. Integrating LLMs into recsys
1.1. LLM-augmented recommenders
• Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendations
- arxiv.org/abs/2306.08121
I wrote a custom Bluesky feed!
It's a classifier trained my past likes. As a bonus, I got to learn more about recommender systems ("RecSys").
adamwiggins.com/posts/a-blu...
It's a classifier trained my past likes. As a bonus, I got to learn more about recommender systems ("RecSys").
adamwiggins.com/posts/a-blu...
A Bluesky feed for one · Adam Wiggins
I built a custom feed for Bluesky that trains a model on my past interactions. Here's what I learned about RecSys and social media feeds generally.
adamwiggins.com
June 2, 2026 at 2:57 PM
I wrote a custom Bluesky feed!
It's a classifier trained my past likes. As a bonus, I got to learn more about recommender systems ("RecSys").
adamwiggins.com/posts/a-blu...
It's a classifier trained my past likes. As a bonus, I got to learn more about recommender systems ("RecSys").
adamwiggins.com/posts/a-blu...
Infinite Scroll RecSys Disorder (ISRSD) is real
December 18, 2025 at 5:04 AM
Infinite Scroll RecSys Disorder (ISRSD) is real
I need some level of recsys
November 21, 2024 at 3:58 AM
I need some level of recsys
decided to create a RecSys (and adjacent) starter pack with @vickiboykis.com @beeonaposy.bsky.social et al
who am I missing?
go.bsky.app/RGf9BvY
who am I missing?
go.bsky.app/RGf9BvY
October 29, 2024 at 11:39 AM
decided to create a RecSys (and adjacent) starter pack with @vickiboykis.com @beeonaposy.bsky.social et al
who am I missing?
go.bsky.app/RGf9BvY
who am I missing?
go.bsky.app/RGf9BvY
Today we have a LLMs +RecSys talk by Jianling Wang from Google Deepmind at our seminar at @irglasgow.bsky.social
November 25, 2024 at 3:08 PM
Today we have a LLMs +RecSys talk by Jianling Wang from Google Deepmind at our seminar at @irglasgow.bsky.social
some of it like anti-semitism in Japan, can maybe be more explained by recsys things or deliberate copying
July 22, 2025 at 3:25 PM
some of it like anti-semitism in Japan, can maybe be more explained by recsys things or deliberate copying
📌 AutoRecLab: Describe the Experiment, Get the Code!
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
AutoRecLab: Describe the Experiment, Get the Code!
Empirical evaluation is central to recommender-systems (RecSys) research, but turning experimental designs into executable code remains a manual and error-prone task. We present AutoRecLab, a Python-based autonomous RecSys lab that automates RecSys experiments from natural-language prompts. Given a…
arxiv.org
September 22, 2026 at 9:54 PM
I love the term "infinite scroll recsys disorder"
December 9, 2025 at 3:58 PM
I love the term "infinite scroll recsys disorder"
Habemus #Dagstuhl Report!
#recsys #evaluation
edited by @evazangerle.bsky.social , @alansaid.com , and yours truly - and written by many, many brilliant minds in our #recsys community.
#recsys #evaluation
edited by @evazangerle.bsky.social , @alansaid.com , and yours truly - and written by many, many brilliant minds in our #recsys community.
And we have a #Dagstuhl Report!
Evaluation Perspectives of #RecSys. Edited by @christinebauer.bsky.social, @evazangerle.bsky.social, and myself. Written by a whole host of fantastic #recsys people. Too many to mention (pic here www.dagstuhl.de/en/seminars/...)
drops.dagstuhl.de/entities/doc...
Evaluation Perspectives of #RecSys. Edited by @christinebauer.bsky.social, @evazangerle.bsky.social, and myself. Written by a whole host of fantastic #recsys people. Too many to mention (pic here www.dagstuhl.de/en/seminars/...)
drops.dagstuhl.de/entities/doc...
Evaluation Perspectives of Recommender Systems: Driving Research and Education (Dagstuhl Seminar 24211)
drops.dagstuhl.de
November 26, 2024 at 9:51 PM
Habemus #Dagstuhl Report!
#recsys #evaluation
edited by @evazangerle.bsky.social , @alansaid.com , and yours truly - and written by many, many brilliant minds in our #recsys community.
#recsys #evaluation
edited by @evazangerle.bsky.social , @alansaid.com , and yours truly - and written by many, many brilliant minds in our #recsys community.
📌 AutoRecLab: Describe the Experiment, Get the Code!
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
AutoRecLab: Describe the Experiment, Get the Code!
Empirical evaluation is central to recommender-systems (RecSys) research, but turning experimental designs into executable code remains a manual and error-prone task. We present AutoRecLab, a Python-based autonomous RecSys lab that automates RecSys experiments from natural-language prompts. Given a…
arxiv.org
September 22, 2026 at 8:07 PM
everyone with higher decoupling than me is an insulated quokka tech bro with no conscience
everyone with lower decoupling than me shamefully outsourced their worldview-forming to the zuckerberg recsys braindemon
everyone with lower decoupling than me shamefully outsourced their worldview-forming to the zuckerberg recsys braindemon
March 10, 2026 at 10:40 PM
everyone with higher decoupling than me is an insulated quokka tech bro with no conscience
everyone with lower decoupling than me shamefully outsourced their worldview-forming to the zuckerberg recsys braindemon
everyone with lower decoupling than me shamefully outsourced their worldview-forming to the zuckerberg recsys braindemon
📌 AutoRecLab: Describe the Experiment, Get the Code!
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
Moritz Baumgart et al.
#arXiv #cs.AI #cs.IR #cs.LG
AutoRecLab: Describe the Experiment, Get the Code!
Empirical evaluation is central to recommender-systems (RecSys) research, but turning experimental designs into executable code remains a manual and error-prone task. We present AutoRecLab, a Python-based autonomous RecSys lab that automates RecSys experiments from natural-language prompts. Given a…
arxiv.org
September 21, 2026 at 8:55 PM
Users have no idea what they want and really don't know what they're GOING to want (for recsys) which leaves devs sniffing around for revealed preferences like a pack of hyenas
November 23, 2025 at 8:54 PM
Users have no idea what they want and really don't know what they're GOING to want (for recsys) which leaves devs sniffing around for revealed preferences like a pack of hyenas
Its a shame that the twitter For You algorithm wasnt released in full. Just maybe it could have been better than Discover github.com/igorbrigadir...
GitHub - igorbrigadir/awesome-twitter-algo: The release of the Twitter algorithm, annotated for recsys
The release of the Twitter algorithm, annotated for recsys - igorbrigadir/awesome-twitter-algo
github.com
January 27, 2025 at 5:13 AM
Its a shame that the twitter For You algorithm wasnt released in full. Just maybe it could have been better than Discover github.com/igorbrigadir...
Going to start a bespoke recsys that pairs blog posts like wine tastings. For your consideration, the first pairing, read these back to back.
One: www.seangoedecke.com/how-to-ship/
Two: ludic.mataroa.blog/blog/why-i-w...
One: www.seangoedecke.com/how-to-ship/
Two: ludic.mataroa.blog/blog/why-i-w...
November 12, 2024 at 12:12 PM
Going to start a bespoke recsys that pairs blog posts like wine tastings. For your consideration, the first pairing, read these back to back.
One: www.seangoedecke.com/how-to-ship/
Two: ludic.mataroa.blog/blog/why-i-w...
One: www.seangoedecke.com/how-to-ship/
Two: ludic.mataroa.blog/blog/why-i-w...
If any of you happen to know (monetized) content creators (TikTok, Instagram, or YouTube) please share! Recruitment for a research study. And for more information: that-recsys-lab.net/home/project...
December 15, 2025 at 8:51 PM
If any of you happen to know (monetized) content creators (TikTok, Instagram, or YouTube) please share! Recruitment for a research study. And for more information: that-recsys-lab.net/home/project...
It would be so cool for some company with the requisite recsys data to unpack it into an infinite library you could walk through.
If you plot the curve of the Dewey Decimal System through embedding space, I wonder how continuous it is.
May 18, 2026 at 3:28 PM
It would be so cool for some company with the requisite recsys data to unpack it into an infinite library you could walk through.
I can see more of the value for facebook and others, where better generative models unlock advertising features you can sell, and more analysis/better recsys. I dont see that business case for XAi
June 24, 2026 at 1:18 PM
I can see more of the value for facebook and others, where better generative models unlock advertising features you can sell, and more analysis/better recsys. I dont see that business case for XAi
A few months ago I went on a recsys dive and what I came away with was a belief that they don’t really know how to measure any other metric with confidence and so it is the unavoidable default
November 23, 2025 at 8:17 PM
A few months ago I went on a recsys dive and what I came away with was a belief that they don’t really know how to measure any other metric with confidence and so it is the unavoidable default
Is it normal for arXiv submissions to be "on hold" for nearly a week now? Our forthcoming RecSys Research & Practice Note has been stuck in the queue since last week 😢.
September 24, 2026 at 2:49 PM
Is it normal for arXiv submissions to be "on hold" for nearly a week now? Our forthcoming RecSys Research & Practice Note has been stuck in the queue since last week 😢.
In the #Beyond Algorithms: Reclaiming the Interdisciplinary Roots of Recommender Systems workshop @recsys.bsky.social 2025, we're calling for bold ideas, critical reflections & work reconnecting RS research with its roots beyondrecsys.github.io @christinebauer.bsky.social @alansaid.com
BEYOND Workshop@RecSys 2025
Website for the BEYOND Workshop@RecSys.
beyondrecsys.github.io
June 10, 2025 at 4:37 AM
In the #Beyond Algorithms: Reclaiming the Interdisciplinary Roots of Recommender Systems workshop @recsys.bsky.social 2025, we're calling for bold ideas, critical reflections & work reconnecting RS research with its roots beyondrecsys.github.io @christinebauer.bsky.social @alansaid.com