LIME-Rec, a non-autoregressive recommender approach using sequential, collaborative, and offline semantic signals, is proposed as an alternative to serving-time LM inference, achieving top reported means…

#RecommenderSystems #LanguageModels #InformationRetrieval
https://arxiv.org/abs/2608.01260

New arXiv work shows they can be poisoned with under 1% synthetic interactions, significantly boosting malicious targets via decoding-aligned prefixes. Worth considering if you build or evaluate…

#GenerativeAI #PromptInjection #RecommenderSystems #AIResearch
https://arxiv.org/abs/2610.14487

New research introduces STATIC, a method that flattens prefix trees into a compressed sparse row matrix to enable efficient constrained decoding for LLM-based generative retrieval on TPUs and GPUs. The approach…

#AIInfrastructure #GPUs #LLMs #RecommenderSystems
https://arxiv.org/abs/2602.22647

CutBCE offers a fused, hardware-accelerated BCE loss and gradient operator in JAX and Pallas, tackling OOM issues in large-vocabulary multi-label recommendation without changing the math. Useful for scaling sequential…

#AI #MachineLearning #RecommenderSystems #JAX
https://arxiv.org/abs/2610.05559

Yandex’s new Sona recommender blends dual attention to weigh both historic and fresh user signals—boosting relevance like never before. Curious how it balances past & recent events? Dive into the research! #YandexSona #DualAttention #RecommenderSystems

🔗 aidailypost.com/news/yandex-...

New paper "ImbalancE" proposes an inference-time latent search method to fix degree imbalance in knowledge graph link prediction, targeting hard cases where the target entity has far fewer connections than…

#AI #KnowledgeGraphs #LinkPrediction #RecommenderSystems
https://arxiv.org/abs/2609.36996

T-RoPE introduces a time-aware variant of Rotary Position Embedding for sequential generative recommendation, using timestamps instead of interaction indices to better capture elapsed time…

#RecommenderSystems #TransformerAI #SequentialRecommendation #GenerativeAI
https://arxiv.org/abs/2609.30576

Spotify’s Taste Profile lets Premium users fine-tune the AI behind Discover Weekly, Wrapped & more. #Spotify #TasteProfile #AI #RecommenderSystems #Streaming #MusicTech thedailytechfeed.com/spotify-lets...

A Job I Like or a Job I Can Get: Designing Job #RecommenderSystems Using Field Experiments https://d.repec.org/n?u=RePEc:arx:papers:2603.21699&r=&r=exp
"… welfare-optimal RSs rank vacancies by an expected-surplus index, and shows why rankings based solely on utility, #hiring probabilities, or 1/5

📉📈 Measuring Bias in Job #RecommenderSystems: Auditing the Algorithms https://d.repec.org/n?u=RePEc:nbr:nberwo:32889&r=&r=ain

"…recommender systems show different jobs to identical male & female job seekers, though most of the jobs overlap

…women are steered towards jobs that pay less, … 1/2

…results suggest that the consistency of an algorithm’s recommendations over time is key to inducing people to follow algorithmic advice in exploration tasks.

#ExperimentalEcon #RecommenderSystems

via https://fediscience.org/@davdittrich/109895784640199297 2/2

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...

#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"

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...

With @bsky.app starting a video feed it’s a great opportunity for #recommendersystems #machinelearning folks to improve the user experience for users. Approaches include after a user watches a video show
- top videos of similar topics to this video
- videos full watched after watching this video
..

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...
#usaref="/hashtag/usa" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link">#usa #gopref="/hashtag/gop" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link">#gop #fascists/hashtag/fascists" class="hover:underline text-blue-600 dark:text-sky-400 no-card-link">#fascists

👉 Vote 'em Out!

👉 𝗗𝗼𝗻’𝘁 𝗺𝗶𝘀𝘀 𝗼𝘂𝘁! 𝗟𝗲𝗮𝗿𝗻 𝗺𝗼𝗿𝗲 𝗮𝗻𝗱 𝗿𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝘁𝗼𝗱𝗮𝘆: impact.dataschool.nl/events/makin...

#RecommenderSystems #Conference #Collaboration #Algorithms #AI #Ethics

#RecommenderSystems #recommendations #MLOps #TwoTower #recsys #machinelearning

Deadline extended to July 27 for the FRAME Workshop at #RecSys2026 !

We welcome papers on experimental methodology.
Research papers will be reviewed results-blind to emphasize motivation and rigorous experimental design.

remaplab.github.io/frame2026/cfp/

#RecommenderSystems

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

Check the call of the reproducibility track, which is also open to resource, methodology and reflective papers!

#recsys #recommendersystems

🧐This work didn’t come out of a vacuum. I built it on top of insights from Amazon Web Services (AWS), Recombee, Spotify, Netflix, and more, complemented with a dash of my own experience/salt🧂

#RecommenderSystems #MachineLearning #DataScience #Personalization

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: https://getnews.me/llm-explanations-improve-transparency-in-recommender-systems/ #recommendersystems #llm

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. https://getnews.me/semanticshield-llm-powered-audits-reveal-shilling-attacks-in-recommender-systems/ #semanticshield #recommendersystems

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