NEWVectors or files. Pick a path.Start →
    Back to DiagramsRetrieval Mechanics

    The Query Is the Biggest Knob

    "chair" and "worn leather armchair, bright room" against the same 118 image index, same retriever, no filters. Zero overlap between the two top fives. Nothing in the configuration changed.

    A portrait diagram in two rails. The headline reads The query is, the biggest knob, with the subline: no configuration changed between these two rails, only the sentence did. The search bar holds the word chair, annotated: nothing else was changed. Rail A, badged broad with the note one word, shows five modern chairs in rank order at 872 dollars 0.098, 515 dollars 0.089, 924 dollars 0.088, 180 dollars 0.087 and 446 dollars 0.087, chip 5 of 5. Rail B, badged specific with the query worn leather armchair, bright room and the note a described scene, shows five different photos: worn leather armchairs at 496 dollars 0.072, 938 dollars 0.071 and 489 dollars 0.069, a red velvet armchair in a bright interior at 451 dollars 0.067, and a dark textured armchair at 606 dollars 0.063; its chip reads a different five. A bold line states that your biggest retrieval knob is the query itself. A code window shows the execute request with two highlighted inputs lines, query chair for A and query worn leather armchair, bright room for B, then filters null and skip_cache true. The footer prints mxp.co/d/query-specificity.
    "chair" and "worn leather armchair, bright room" against the same 118 image index, same retriever, no filters. Zero overlap between the two top fives. Nothing in the configuration changed.

    Two searches against the same 118 image index. Same retriever, same embeddings, same filters, which is to say none. The one word query chair returns generic modern chairs. The described scene, worn leather armchair, bright room, returns worn leather armchairs in bright rooms.

    Zero items overlap between the two top fives.

    Most retrieval tuning happens downstream of the query: top_k, fusion weights, rerankers, filters. All of those matter, and none of them moved here. The largest change in results came from the sentence the index was asked to match.

    The practical reading is that the cheapest improvement available to most search products is a longer query. Help users write one, with suggestions or a describe what you want prompt, or write a better one on their behalf before it reaches the index. Query expansion, rewriting and hypothetical document embeddings are all ways of doing the second thing automatically.

    The scores tell a second story. The broad query scores higher, 0.087 to 0.098, than the specific one, 0.063 to 0.072, because a one word query is close to many images and a described scene is close to few. A higher score means more of the corpus looks like the query. Which query produced the result the user wanted is a separate question, and clicks answer it.

    What is real here: the photos are Pexels images with every photographer credited, the embeddings and ranking came from the live API with the cache skipped on both calls, and the order is untouched. Prices are demo catalog data generated for this corpus.

    Run this on your own data

    Mixpeek turns video, images, audio, and documents in your object storage into searchable, timestamped results through one API.

    Search your own data

    Related diagrams

    A portrait diagram in two rails. The headline reads Filters exclude, what ranking would have shown, with the subline: a pre_filter drops rows before ranking is ever shown to you, useful, and it hides what it removed. A search bar holds the query yellow armchair, annotated: three of the top five are out of stock. Rail A, badged no filter, shows five armchair photos in rank order with prices and scores: 872 dollars at 0.080, 515 dollars at 0.065, 180 dollars at 0.060, 451 dollars at 0.060 and 506 dollars at 0.057; the second, third and fourth are greyed and stamped dropped, and the count chip reads 5 of 5. Rail B, badged pre_filter with in_stock equals true, shows the two survivors at 872 and 506 dollars followed by three new armchairs at 123 dollars 0.048, 446 dollars 0.047 and 606 dollars 0.046; its chip reads 2 kept plus 3 new. A bold line states that three of the top five were out of stock and all three were removed from the response entirely. A dark code window shows the execute request with the filters line in_stock true highlighted, and skip_cache true. The footer prints mxp.co/d/filters-exclude and the words real query, real ranking, mixpeek.

    Filters Exclude What Ranking Would Have Shown

    A portrait diagram in two rails. The headline reads Scope decides, what can be found at all, with the subline: the same question, asked of the whole catalog and then of one category. The search bar holds the query something to keep me warm outside, annotated: the query names no category. Rail A, badged whole catalog with the note semantic reach across every category, shows five photos in rank order: a lit tent at night for 543 dollars at 0.062, a snowy forest scene for 564 dollars at 0.060, a steaming pot on a stove for 444 dollars at 0.059, boots on a textured rug for 203 dollars at 0.052, and a tent under a starry sky for 100 dollars at 0.048; the second, third and fourth are greyed and stamped dropped, and the chip reads 5 of 5. Rail B, badged scoped with category equals tent and the note only one category can answer, shows the two surviving tents followed by three more tents at 592 dollars 0.046, 73 dollars 0.039 and 554 dollars 0.029; its chip reads 2 kept plus 3 new. A bold line states that unscoped, the index answered with tents, backpacks, rugs and pans, and that scope is a decision about what is allowed to be right. A code window shows the execute request with the filters line category tent highlighted. The footer prints mxp.co/d/scope-changes-candidates.

    Scope Decides What Can Be Found at All

    A portrait diagram in two rails. The headline reads A price ceiling, is not a price sort, with the subline: a price ceiling removes the expensive matches, it does not promote the cheap ones by price. The search bar holds the query premium floor lamp, annotated: premium is a word, not a number. Rail A, badged no ceiling with the note best matches, any price, shows five photos in rank order: floor lamps at 729 dollars 0.097, 299 dollars 0.093, 908 dollars 0.092 and 322 dollars 0.089, then an armchair at 451 dollars 0.086; four are greyed and stamped dropped, only the 299 dollar lamp survives, and the chip reads 5 of 5. Rail B, badged pre_filter with price less than 300 and the note best matches under the ceiling, shows the 299 dollar lamp first at 0.093, then lamps at 230 dollars 0.071, 104 dollars 0.068, 129 dollars 0.063 and 38 dollars 0.059; its chip reads 1 kept plus 4 new. A bold line states that rank order inside the filter is still semantic, so the cheapest item lands fourth. A code window shows the execute request with the filters line price lt 300 highlighted. The footer prints mxp.co/d/numeric-filter-reranks.

    A Price Ceiling Cuts the Set and Leaves the Order Semantic