Free tools Windows power users keep installed
One-click scans. No signup required.
Netflix says its MAPS system uses image, video, audio and text embeddings to personalize artwork and video previews before a new asset has built up much interaction history. In a company-authored account published August 28, 2026, Netflix reports that combining content signals with pooled interaction data improved its artwork and preview systems in tests; it does not disclose the online effect sizes.
Why new artwork and previews are hard to personalize
Many personalization models learn from how members respond to known assets, such as a particular poster or video preview. That creates a cold-start problem: a newly created asset has little or no interaction history, so an asset-ID model has limited evidence about which members might prefer it. Netflix says it historically responded by increasing exploration or falling back on popularity heuristics until more interactions accumulated.
As an Amazon Associate I earn from qualifying purchases.
Content embeddings offer another signal. Instead of representing an asset only by an ID learned from past behavior, a model can use information about its visual appearance, audio and text. Netflix frames the practical question as: “But which image or video preview of Squid Game should we show you?” The same issue arises immediately after a title launches, before the system has enough response data to choose an asset for each member.
How MAPS personalizes artwork across five canvases
Combining visual content with interaction history
For artwork, Netflix says it encodes each image with pretrained CLIP, producing a 768-dimensional image embedding, and concatenates that vector with the asset’s learned ID embedding. A multilayer perceptron (MLP) then produces an asset representation that is scored against a member. The ID carries learned interaction history where it exists; the image vector gives the model content information that can also be used for a newly created asset.
#1 Best Overall
- HD streaming made simple: With America’s number 1 TV streaming platform,* exploring popular apps—plus tons of free movies, shows, and live TV—is as easy as it is fun. *Based on hours streamed—Hypothesis Group
- Compact without compromises: The sleek design of Roku Streaming Stick won’t block neighboring HDMI ports, and it even powers from your TV alone, plugging into the back and staying out of sight. No wall outlet, no extra cords, no clutter.
- No more juggling remotes: Power up your TV, adjust the volume, and control your Roku device with one remote. Use your voice to quickly search, play entertainment, and more.
- Shows on the go: Take your TV to-go when traveling—without needing to log into someone else’s device.
- TV, simplified: With setup that only takes minutes, a simple-to-navigate Home Screen, and an uncluttered remote control that does all you need—Roku makes it easier to watch the TV you love.
Netflix describes using one model across five artwork canvases: billboard, vertical-box, horizontal-panel, short-panel and landscape-panel. Its rationale is that CLIP is relatively invariant to crop, resize and aspect ratio, making it possible to share interaction signals across differently shaped surfaces. Netflix says an affinity learned on a high-traffic canvas can inform artwork selection on a sparse one, and reports the largest gains on canvases with the least interaction data.
Pooling canvases without letting volume dominate
Pooling examples across canvases creates an imbalance: surfaces differ in impression volume and in the kinds of interactions they generate. Netflix says it weights training examples according to the long-term reward assigned to the interaction type. The company describes this as a way to avoid choosing a separate manual weight for each canvas and to reduce the influence of high-volume, short-term actions.
Rank #2
- Advanced 4K streaming - Elevate your entertainment with the next generation of our best-selling 4K stick, with improved streaming performance optimized for 4K TVs.
- The newest Fire TV experience (2026) – Our biggest update to Fire TV has a new, modern design that gets you to your entertainment fast. Browse dedicated content categories, pin more of your favorite apps, and get personalized recommendations from Alexa+. Spend less time scrolling, and more time watching.
- Cloud gaming, no console required – Stream Call of Duty: Black Ops 7, Hogwarts Legacy, Outer Worlds 2, Ninja Gaiden 4, and hundreds of games on your Fire TV Stick 4K Select with Xbox Game Pass and Luna via cloud gaming. Xbox Game Pass subscription and compatible controller required. Each sold separately.
- Smarter picks with Alexa+ – Getting to what you love has never been easier. Press the voice remote button and talk naturally to find what to watch across your apps, manage your smart home, or dive into virtually any topic.
- Wi-Fi 6 support - Enjoy smooth 4K streaming, even when other devices are connected to your router.
What Netflix’s artwork ablation compared
| Version | Training design | Netflix’s reported result |
|---|---|---|
| V1 | Image embeddings, with separate models for each canvas | Did not produce a statistically significant online lift in the reported ablation. |
| V2 | One pooled model across canvases, using IDs only | Did not produce a statistically significant online lift in the reported ablation. |
| V3 | One pooled model across canvases, using IDs and image embeddings | Produced a statistically significant online lift in the reported ablation; Netflix says the combined approach runs in production. |
The comparison suggests that, in Netflix’s reported test, neither image embeddings in separate models nor pooled training with IDs alone was enough to produce a significant lift. The version combining pooled training and image content did. Netflix reports a 5.691% short-panel offline lift for V3 over the combined V1 and V2 lifts. That is an offline figure, not an online lift. The article says changes within ±1% on its reported offline metric are not significant.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Why the TV redesign made sparse-canvas performance important
Netflix says a redesigned TV home screen made short-panel the dominant artwork canvas, even though short-panel had little historical interaction data. The company reports shipping V3 ahead of the redesign and evaluating it in a one-month holdback A/B test. Netflix says the model significantly improved its core discovery metric and streaming hours in that test, but does not publish the online effect sizes.
Rank #3
- 4K streaming made simple:With America’s number 1 TV streaming platform,* exploring popular apps—plus tons of free movies, shows, and live TV—is as easy as it is fun. *Based on hours streamed—Hypothesis Group
- 4K picture quality: With Roku Streaming Stick Plus, watch your favorites with brilliant 4K picture and vivid HDR color.
- Compact without compromises: Our sleek design won’t block neighboring HDMI ports, and it even powers from your TV alone, plugging into the back and staying out of sight. No wall outlet, no extra cords, no clutter.
- No more juggling remotes: Power up your TV, adjust the volume, and control your Roku device with one remote. Use your voice to quickly search, play entertainment, and more.
- Shows on the go: Take your TV to-go when traveling—without needing to log into someone else’s device.
How search adds the query to artwork ranking
For search results, Netflix says it combines the artwork model’s member-personalization score with cosine similarity between a CLIP text embedding of the query and the CLIP image embedding of each asset. An α mixing weight is tuned in online A/B tests. This gives ranking a way to account for both the member’s general taste and the specific query, rather than relying on either signal alone.
How MediaFM represents video previews
From visual-only frames to multimodal shots
Netflix’s earlier content-aware approach, SeqCLIP, encoded frames from a preview with CLIP and averaged the frame representations. Netflix describes that approach as capturing visual appearance while missing audio and dialogue. Its MediaFM model combines three per-shot signals: visual SeqCLIP features, a pretrained speech/audio embedding and captions encoded by a large-scale text model. Netflix says MediaFM was trained at a scale of 80 million shots.
Rank #4
- The Google TV Streamer (4K) delivers your favorite entertainment quickly, easily, and personalized to you[1,2]
- HDMI 2.1 cable required (sold separately)
- See movies and TV shows from all your services right from your home screen[2]; and find new things to watch with tailored recommendations for everyone in your home based on their interests and viewing habits
- Watch live TV and access over 800 free channels from Pluto TV, Tubi, and more[3]; if you find an interesting show or movie on your TV, mobile app, or Google search, you can easily add it to your watchlist, so it’s ready when you are[2]
- Up to 4K HDR with Dolby Vision delivers captivating, true-to-life detail[4]; and you can connect speakers that support Dolby Atmos for more immersive 3D sound
Reported comparison of preview embeddings
| Approach | Signals used | Evidence and reported status |
|---|---|---|
| ID-only | Asset interaction history, without the visual or multimodal content signals described for the other approaches | Online baseline in Netflix’s reported comparison. |
| SeqCLIP | Visual preview frames encoded with CLIP and averaged | Compared offline and in the online test; Netflix reports it ranked above ID-only. |
| MediaFM | Visual features, speech/audio embeddings and captions encoded as text | Netflix reports it ranked above SeqCLIP and ID-only, delivered a statistically significant lift in its core streaming metric over ID-only, and became the default video-preview embedding across platforms. |
Netflix says it evaluated the three approaches offline with inverse propensity scoring and in a five-week online A/B test across device platforms. The reported ordering was MediaFM, then SeqCLIP, then ID-only, with the largest gains on TV. The MAPS article does not disclose the online lift’s size, so the reported ranking and significance should not be read as a quantified effect for every platform.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →How Netflix screens models before online tests
Inverse propensity scoring
Netflix describes its offline metric as inverse propensity scoring (IPS) on a dedicated exploration slice. Because assets in that slice are selected randomly, their serving propensities are known. IPS reweights each observation by the inverse of its serving propensity to account for the selection probability. Netflix says a candidate must beat the production baseline offline before it receives A/B-test traffic.
Best Value
- Stunning 4K and Dolby Vision streaming made simple: With America’s number 1 TV streaming platform,* exploring popular apps—plus tons of free movies, shows, and live TV—is as easy as it is fun. *Based on hours streamed—Hypothesis Group
- Breathtaking picture quality: Stunningly sharp 4K picture brings out rich detail in your entertainment with four times the resolution of HD. Watch as colors pop off your screen and enjoy lifelike clarity with Dolby Vision and HDR10+.
- Seamless streaming for any room: With Roku Streaming Stick 4K, watch your favorite entertainment on any TV in the house, even in rooms farther from your router thanks to the long-range Wi-Fi receiver.
- Shows on the go: Take your TV to-go when traveling—without needing to log into someone else’s device.
- Compact without compromises: Our sleek design won’t block neighboring HDMI ports, so you can switch from streaming to gaming with ease. Plus, it’s designed to stay hidden behind your TV, keeping wires neatly out of sight
A lower-cost proxy for comparing embeddings
Full offline evaluation and online testing are more costly than a quick model screen. Netflix describes a linear probe that predicts, from an embedding alone, the asset that wins for each title after popularity is debiased. It uses exploration data and propensity adjustment to label that winner. The probe helps narrow the candidates before full IPS evaluation and online testing; Netflix says probe accuracy, IPS and online results all ranked MediaFM ahead of SeqCLIP in its reported comparison. Netflix also says it now uses the probe to screen each new MediaFM checkpoint.
Why shared embedding infrastructure matters
Netflix describes its Embedding Store as part of its AI Platform. It holds dense embeddings for titles, games, member profiles and multimedia assets, and serves the same embeddings to training and online inference. That shared path is intended to keep the representation used when a model is trained consistent with the one available when it makes a recommendation.
The store also separates embedding-model updates from the personalization models that consume them. Netflix says a new embedding can be registered, backfilled and validated independently, then made available through configuration without requiring downstream code changes. That makes it possible to update a content representation without rebuilding every system that uses it.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallWhat Netflix’s reported results do—and do not—establish
The MAPS account is Netflix’s own description of its methods, experiments and deployment, not an independent evaluation. It reports significant test outcomes and relative model rankings, but withholds the online effect sizes for the artwork and preview results discussed here. Its 5.691% figure is explicitly an offline short-panel comparison; it should not be substituted for an online result.
A related arXiv record, submitted August 18, 2026, is titled “Multimedia Asset Personalization via Multimodal Embeddings at Netflix” and has a different author list from the eight-author Netflix TechBlog article. It is a separate publication record, not simply another title for the August 28 blog post.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
