The ACM RecSys 2026 Tutorials Track is now open for submissions!

We welcome proposals on recommender systems topics, especially tutorials addressing evaluation, deployment, feedback loops, and real-world constraints.

#recsys #recommendersystems #RS_c

Krea.ai is hiring an ML Engineer to architect its personalization & recommendation stack from the ground up, blending recommendation systems with generative AI. 📍San Francisco. Tech: Python, PyTorch, JAX, deep learning #MachineLearning #AI #RecommenderSystems #Hiring aihackerjobs.com/company/krea...

The modular architecture of WarpRec. Five decoupled modules manage the end-to-end recommendation lifecycle, from data ingestion and processing to model training and evaluation. An Application Layer exposes the recommender through a REST API and MCP agentic interface.

Future recommendation infrastructures must integrate evaluation protocols, fairness metrics, and reproducible pipelines as first-class design principles—not afterthoughts.
The paper “WarpRec” proposes a framework that unifies academic rigor with industrial-scale […]

[Original post on det.social]

Opinion: Banning under-16s from social media is a half-measure. We should ban toxic algorithms Australia is trying to ban teens from social media, but that only scratches the surface as a solution to the problems tech companies are causing us, writes Killian Mangan.

Please support the ban! @fine-gael.bsky.social @fiannafailparty.bsky.social @greenparty.ie @aontuie.bsky.social

www.thejournal.ie/prev/6901718...

#Ireland #IrishPolitics #Algorithms #RecommenderSystems #ToxicAlgorithms #SocialMedia #EUPolitics #Democracy #MentalHealth #Tech #TechRegulation

"Banning #socialmedia for young people will ignore the incredibly harmful societal effects of modern social media for most of the population…
The most immediate solution is to ban companies from using #recommendersystems entirely (outside a few specific cases); that would restore our freedom to […]

#LLM vs #RecommenderSystems (Part 2/3) 👉 recsysml.substack.com/p/recsys-ret... is about Retrieval.
And this is where the analogy with LLMs becomes surprisingly tight.

Core idea:
Retrieval in recommender systems plays the same role as pretraining in LLMs.

We show clustering > "Semantic IDs"

Opinion: Banning under-16s from social media is a half-measure. We should ban toxic algorithms Australia is trying to ban teens from social media, but that only scratches the surface as a solution to the problems tech companies are causing us, writes Killian Mangan.

My article on @TheJournal , calling for toxic algorithms on social media to be banned!

#socialmedia #australia #ireland #eu #socialmediaban #bansocialmedia #recommendersystems #democracy #farright #extremism #fascism #online #iris...
#usa #gop #fascists

👉 Vote 'em Out!

Opinion: Banning under-16s from social media is a half-measure. We should ban toxic algorithms Australia is trying to ban teens from social media, but that only scratches the surface as a solution to the problems tech companies are causing us, writes Killian Mangan.

My article on @thejournal.ie calling for toxic algorithms on social media to be banned!

jrnl.ie/6901718

#socialmedia #australia #ireland #eu #socialmediaban #bansocialmedia #recommendersystems #democracy #farright #extremism #fascism #irish #irishpolitics #eupolitics #europe #toxicalgorithms

Ranking Models explained: Deep Dive into RecSys Architecture (Features, Embeddings, & Attention) Watch now (18 mins) | What happens post retrieval in a recommender system

In this three part series I compare #LLM and #RecommenderSystems to show the gaps and opportunities. They are surprisingly fewer than one would think.

Part 1 open.substack.com/pub/recsysml...

At a time when the far-right, aided by toxic social media algorithms and billionaire media, are selling hate to billions, the Left has seen a huge resurgence based on a radical hope.

#FarRight #Algorithms #RecommenderSystems #Hate

A screenshot of an email from ResearchGate that says, We found a free webinar that matches your interests. The body of the message says, A reliable CNS safety assessment method. CNS safety risks often go undetected in the GLP talk studies because animal models fail to predict human neurological effects. CNS-3D brain organoids are 7.4 times more accurate than animal models offering a more reliable CNS safety assessment. It has to be said that I am not a neurological or biological scientist. I am in fact a technology in society, in urban and learning contexts academic. So in fact have absolutely nothing to do with this field of study.

I love it when recommender systems are so chronically off that it just confirms the coming automated dystopian future we have built will be 90% Brazil and 10% LOTF.

#recommendersystems #researchgate #academia #academicchatter

Scalable LinUCB: Low-Rank Design Matrix Updates for Recommenders with
Large Action Spaces
Evgenia Shustova, Evgeny Frolov et al.
Paper
Details
#ScalableLinUCB #RecommenderSystems #LargeActionSpaces

Glasgow Information Retrieval Group | LinkedIn Glasgow Information Retrieval Group | 298 followers on LinkedIn. Founded in 1986, the Glasgow IR Group has been at the forefront of Research & Development in search and recommendation.

📢 We're also now on LinkedIn!

Follow the Glasgow Information Retrieval Group for updates on IR research, @irglasgow.bsky.social activities, events, and collaborations:

🔗 linkedin.com/company/glasgow-information-retrieval-group

#InformationRetrieval #IR #AI #recsys #RecommenderSystems #Glasgow

In der Schweiz übrigends auch ein Problem, ich wollte nur mal dran erinnern. #recommendersystems #filterbubble

LLM Explanations Improve Transparency in Recommender Systems

A study of 326 participants found large language models can turn matrix-factorization recommendations into clear explanations that boost perceived transparency and trust. Read more: getnews.me/llm-explanations-improve... #recommendersystems #llm

SemanticShield: LLM-Powered Audits Reveal Shilling Attacks in Recommender Systems

SemanticShield uses a two-stage LLM detector that audits item descriptions in real-time. The paper was submitted in September 2025 and the code is on GitHub. getnews.me/semanticshield-llm-power... #semanticshield #recommendersystems

RecInter: Interaction‑Centric Agent Simulation for Dynamic Recommenders

RecInter, an agent‑based simulation platform for recommender systems presented at EMNLP 2025, lets user actions instantly update item attributes; code is on GitHub. getnews.me/recinter-interaction-cen... #recommendersystems #recinter

Reciprocal Retrieval-Generation Boosts Conversational Recommender Systems

ReGeS links retrieval and generation in a reciprocal loop to sharpen intent extraction and lower hallucinations in conversational recommender systems; its code is on GitHub. getnews.me/reciprocal-retrieval-gen... #recommendersystems #reges

Benchmarking LLM‑Based Evolutionary Algorithms for Recommender Systems

RSBench, a new benchmark for LLM‑driven evolutionary algorithms, evaluates prompts on accuracy, diversity and fairness, with three algorithms showing distinct Pareto fronts. getnews.me/benchmarking-llm-based-e... #rsbench #recommendersystems #llm

Intelligent Algorithm Selection Boosts Recommender System Accuracy

Including algorithm descriptors raised the meta‑learner’s NDCG@10 to 0.143 (11.7% over the 0.128 baseline) and lifted Top‑1 selection accuracy by 16.1%. Read more: getnews.me/intelligent-algorithm-se... #recommendersystems #meta‑learning

Side‑Feature‑Aware Fake Profiles Threaten Recommender Systems

A new study extends Leg‑UP to generate side‑feature‑aware fake profiles, achieving stronger attack performance and low detection rates on benchmark recommender datasets. Read more: getnews.me/side-feature-aware-fake-... #recommendersystems #shillingattack

#INTERSPEECH2025 #ConversationalAI #SpeechProcessing #RecommenderSystems

Benchmark Aligns Recommender System Unlearning with Real‑World Needs

New benchmark for recommender‑system unlearning shows a custom algorithm can delete data with latency of only a few seconds. Posted 18 September 2025. Read more: getnews.me/benchmark-aligns-recomme... #recommendersystems #unlearning

Key Factors in Using LLMs for Recommender Feature Extraction

RecXplore, a modular LLM‑feature framework, boosted sequential recommendation performance by up to 18.7% in NDCG@5 and 12.7% in HR@5 on four public benchmarks. Read more: getnews.me/key-factors-in-using-llm... #recommendersystems #llm #featureextraction

Model‑Agnostic Post‑Hoc Explainability Improves Recommender Systems

Deletion diagnostics measures influence by comparing performance with and without observation. It was applied to Neural Collaborative Filtering on the MovieLens dataset. getnews.me/model-agnostic-post-hoc-... #recommendersystems #explainability

Low‑Rank Adapter Fine‑Tuning of Small Language Models for User Behavior

Researchers use low‑rank adapters to fine‑tune small language models as user simulators, handling millions of personas with far less compute than large LLMs. Read more: getnews.me/low-rank-adapter-fine-tu... #recommendersystems #lowrankadapters

Recommender System Evaluation (Part 3): Real-World Deployment - When Rubber Meets the Road Go beyond offline accuracy to truly evaluate your recommender system. This guide covers A/B testing, conversion funnels, fairness, and the business metrics that drive real-world success and retention.

Offline metrics vs. real-world impact for recommender systems? 🤔 Part 3 dives into bridging the gap with A/B testing, business value, & fairness! It's more than just accuracy. Learn how to truly evaluate. 👇 fanyangmeng.blog/recommender-... #RecommenderSystems #MLEvaluation

Recommender System Evaluation (Part 2): Beyond Accuracy - The User Experience Dimension Go beyond accuracy. Learn to evaluate recommender systems with key UX metrics like diversity, novelty, and serendipity to build systems users truly love.

Is your recommender system *just* accurate? 🤔 True value lies in UX metrics! Explore diversity, coverage, & serendipity to build systems users truly love. Learn more: 👇
fanyangmeng.blog/recommender-system-evalu... #RecommenderSystems #UX

Recommender System Evaluation (Part 1): The Foundation - Accuracy and Ranking Metrics Learn essential recommender system evaluation metrics beyond accuracy: NDCG, Precision@K, MAP, and RMSE. Master ranking quality measurement to build recommendation systems users actually love.

Recommender systems: Is your model just "accurate" or truly useful? 🤔 Part 1 explores why MAE/RMSE aren't enough. Discover crucial ranking metrics like NDCG for better user experience! 👇 fanyangmeng.blog/recommender-system-evalu... #RecommenderSystems #MachineLearning #DataScience