Hand-Built Recommender Loses to the 'Most Popular' Baseline Half the Time
dima806_dima · x · 2026-11-08
An 8-minute Medium post by Dima Iakubovskyi runs a humbling experiment: after implementing collaborative filtering and matrix factorization recommenders by hand, he benchmarks them against the naive 'most popular items' baseline.
The verdict is in the title — the fancy recommender loses half the time. A useful reminder to always run simple baselines before reaching for complex models; much of the supposed 'intelligence' isn't there.
More from Research
- Fully Hidden Test Sets Make Benchmarks Harder to Adopt, Researcher Notes — srchvrs · 2026-11-09
- Scott Alexander vs Pinker scaling spat: LLM scaling laws are inverse power laws, never exponential — sebkrier · 2026-11-09
- YC Paper Club: AI for Biology, From Reading Data to Designing Living Systems — ycombinator · 2026-11-08
- NYT: OpenAI results upend a key tenet of computer science — P vs NP — _akpiper · 2026-11-08
- Running Qwen3.5-0.8B in pure Assembly: 6.51 tok/s on one CPU thread — suayptalha · 2026-11-08
- GPT-5.2 Pro solves COLT 2022 open problem on accelerated L1-regularized PageRank — kfountou · 2026-11-08