The way I understand this works is that the researchers found a clever architectural hack to stop AI from hoarding memory when reading long documents.
Normally, when an AI transcribes a 100 page PDF, it tries to remember every single word it has already ingested. This short-term memory (the KV cache) grows linearly O(N) until the model runs out of VRAM and crashes (or caps it) To avoid this, developers are forced to build janky code that chops PDFs into individual pages, processes them one by one, and glues the text back together.
Unlimited OCR uses Reference Sliding Window Attention (R-SWA) to split the AI's focus into two paths:
Global Reference: The AI keeps full, uncompromised sight of the original document image so it never loses context.
Local Generation: The AI restricts its memory of its own typed text to a tight, moving window (like the last 128 words) and safely forgets the rest.
Will be very interesting for local AI and can’t wait to see what the community builds and extends with it!
_puk 5 hours ago [-]
This hits a sweet spot I think for conversations too.
I've been playing (for quite a while) on trying to encapsulate long running conversations.
You have the overriding context, facts that don't change very often at all. The participants names, their backgrounds etc.
Then you have some very fine grained facts (what they ate for breakfast this morning) which might be useful right now, but are irrelevant outside of a general trend over the longer term.
When trying to reconstruct a conversation you really need to find the right balance without pulling in everything that has ever been discussed.
This definitely is worth further investigation.
timwis 4 hours ago [-]
Can you say more about how this applies to long-running conversations? I've been thinking about them as well, but can't write wrap my head around how this would be better than (or even different to) standard compaction.
dominotw 3 hours ago [-]
standard compactions doesnt really distinguish between long term vs short term ephemeral facts ?
timwis 3 hours ago [-]
Forgive me if I'm being naive, but can't you just tweak the compaction prompt to differentiate? Presumably that's what you would do in the separate prompt anyway, right?
ewild 5 hours ago [-]
This sounds like we are trying to add an LSTM into a transformer
htrp 4 hours ago [-]
Sepp would like a word
storywatch 4 hours ago [-]
Haven't read the full paper but thr local generation window is a little small, especially since image inputs are especially token heavy. Depending on where the local attention layer is located, it would be nicer if it's bigger e.g. 4096 words at least.
MattRogish 4 hours ago [-]
I do OCR of images, and that's exactly what I do. I take one big image and slice it into many smaller ones, and send those to the LLM. Perfect every time, unlike using the whole image which resulted in hot garbage.
freefaler 3 hours ago [-]
It works with relatively good scans, when there are bad/skewed scans and especially something with many label/value pairs, that aren't nicely tucked inside sentences, the more context you have, the more you can find the correct words and fix the errors.
There is a whole class of tricky documents. A decent (if you ignore the marketing bias) post about this problem can be found here:
How do you know where to slice an image? What if you slice an image mid-word?
MattRogish 1 hours ago [-]
I calculate* the appropriate overlap and the slicer overlaps a certain amount of the previous slice. There is some post-processing assembly required, but it's trivial.
[*] SWAG line height, trial and error to figure out the right amount of overlap given LLM error rates, etc.
ryanisnan 1 hours ago [-]
Interesting. Do you have a uniform data set? E.g. documents of a specific type that you know consistently have similar formats, or is this training something you need to do per-document?
d675 6 hours ago [-]
See, leetcode is useful. As I do this leetcode grind, I’ve been why techniques exist / how they’re used irl. Lots of interesting stuff there
ai_fry_ur_brain 6 hours ago [-]
Who said it wasnt useful, dont listen to those people.
Xevion 6 hours ago [-]
People who are applying to jobs and are tested with LeetCode problems to assess their skill level, despite the two not really being correlated or relevant for the position
galbar 5 hours ago [-]
As someone that gets very annoyed when having to do LeetCode in interviews...
Knowing algorithms, data structures and their memory and time complexities is very relevant for SWE. I've had teammates that didn't understand them and everything was fine until when it wasn't (scaling and performance issues).
Or, as I put it to a teammate: "Would you rather review the PR of someone that understands the difference between a set and a list or the PR of someone who doesn't?". This was after we interviewed a candidate with ~15 YoE, on paper, that didn't know the difference.
elliottcarlson 5 hours ago [-]
> Knowing algorithms, data structures and their memory and time complexities is very relevant for SWE
Agree with this; however knowing how to roll your own BFS/LRU/etc isn't -- in that case I'd rather review the PR of someone who understands how to leverage tested and known implementations than the PR of someone who decided to roll their own.
ai_fry_ur_brain 4 hours ago [-]
Who care's if the leetcode question doesn't relate to the job itself, it shows whether or not the person is willing to put in the work and gives you a glimpse into their ability to reason about hard problems.
peatmoss 6 hours ago [-]
I recently bought a tablet for sheet music, mostly to replace a stack of jazz "Real Books" at jam sessions. And the phone camera scans I made are okay, but fixed in size and have a lot of artifacts. And it would be great to transpose on the fly for e.g. Bb or Eb instruments, but being a scan this is obviously not possible.
I got digging into the state of optical music recognition and came away concluding that music is basically a greenfield for AI wherever you look. Optical music recognition is pretty terrible. AI understanding of music theory is terrible (actually looking at music that is; LLMs do okay at text descriptions of theory concepts where you can imagine some online texts making it in).
I think the issue is that we still don't have great digital formats that encode the dots on paper that musicians read. Music notation is pretty rich. Midi doesn't capture all of what's needed for symbolic understanding, because it was mostly made for capturing aspects relevant for playback or performance. MusicXML seems to be the closest for a digital format that encodes the information a musician would want, but there aren't great corpora of training data that would connect a MusicXML representation to sheet music images or to audio. I think that's because MusicXML falls short of encoding enough information to engrave music. Tools like MuseScore need to track a bunch of layout information that isn't encodable in MusicXML. Lilypond format is less verbose that MusicXML and contains a bit more information that is useful to the score creators, but most people don't create sheet music in lilypond. (As an aside, Lilypond bums me out with the state of jazz fonts. I hate looking at "legit" scores in jazz context)
I realize this is mildly off topic, but every time I see people making incremental gains on OCR, which to my mind is pretty good, I am reminded of how abysmal OMR is.
kwon-young 5 hours ago [-]
So, the format for musicologist and researcher in music is the MEI format: https://music-encoding.org/ for which the reference engraver is verovio: https://www.verovio.org/index.xhtml
Note that verovio is able to engrave in svg format while keeping a maximum of information from the original mei score, meaning that you can extract enough metadata to create an actual detection dataset for a deep learning model.
This is my horrible hacked up script that will create a coco dataset from a set of scores engraved with verovio: https://github.com/kwon-young/music/blob/main/svg2pl.py
I have published a synthetic music score dataset from this: https://www.kaggle.com/datasets/kwonyoungchoi/trompa-coco/da...
I anyone wants to try and fit a detector on top is welcome :)
To understand why OMR is so neglected is because most people widely underestimate the difficulty of the task.
It has a specific blend of the most extreme shapes combined with an extremely complicated graphical grammar...
peatmoss 4 hours ago [-]
Thank you for this! Both MEI format and the Verovio engraver are news to me. I will check them out.
My first thought was whether MEI format is being added to MuseScore (the sheet music editor I use these days). It looks like it is: https://music-encoding.org/musescore-doc/
As a somewhat related aside, now that the MuseScore people own Hal Leonard and seem to pushing integration with their cloud subscription service, I wonder if they'll see some of these directions as potentially competing with them. I don't think there's anyone who wouldn't love a transposable clean digital version of their Real Books... and if Hal Leonard is in the business of selling Real Books, I can see where good OMR might be a problem for them. I guess piracy of scanned versions is already rampant, so maybe it's a wash.
indiv0 5 hours ago [-]
> music is basically a greenfield for AI wherever you look
AIN'T THAT THE TRUTH.
My girlfriend is studying musicology and she has some physical disabilities that make it difficult for her to write things down sometimes. So I try to help her by writing some AI-powered TTS/OCR/etc. apps here and there. It becomes painfully obvious that music was never considered an important part of any AI training dataset, anywhere.
These days, I'm pleasantly surprised by how well Opus 4.8 understands/explains music theory (as you said). But ask him to transcribe/OCR/OMR some sheet music and he'll confidently give you the MusicXML/Lilypond equivalent of "2 + 2 = horse".
I really hope this ignored area will be swept up with the rest of the rising AI wave, but it's still criminally undervalued.
mejutoco 1 hours ago [-]
> how well Opus 4.8 understands [...] and he'll confidently
I always think of the nun character against AI in Mrs Davis:
> "Don't give it a name. No one calls Facebook Doug. No one calls Twitter Mary Lou. No one calls them anything, because no one uses them anymore. They use it, and it's not a person. It's code. - Mrs Davis
mft_ 11 minutes ago [-]
Eh, humans give names to and/or anthropomorphise lots of things. My partner names all of her cars and bikes; I don't. Isn't it more rational to feel some sort of connection and anthropomorphise a tool with which you can at least have an intelligent conversation, than a simple machine?
peatmoss 4 hours ago [-]
I recently left a job at where I was working with open data producers / providers across a lot of domains. A lot of data is produced and released for free by governments and nonprofits because it's either directly part of the mission, or it's a natural byproduct of the organization's mission. Occasionally, you'd have really great datasets come out of industry / commercial organizations because the data were a byproduct and didn't create a scenario where a data release would create opportunity for competition.
I've been thinking about what kind of organization could be self-sustaining and also produce good music AI training data as a natural byproduct. An ideal arrangement would be something that provided some incentive or benefit to musicians in exchange for their recorded interpretation of sheet music. Soundslice, mentioned by another user, seems to do that. They let both teachers and students upload recordings of music that has been turned into MusicXML. The recordings, paired to those snippets of sheet music, has to be a gold mine. Assuming they have enough users. If they aren't already working on stem separation and automatic transcription, they probably should be. Still, my hope would be to figure out some kind of sustainable model where that dataset could be created and released for open model development...
As a domain, I see AI in music as a boon to human creativity. I am very much a novice jazz improvisor, and a passable amateur technician on the trombone. Human instructors can do a lot for me, but there's a lot that is "grinding it out" repetition, where I think AI could be a huge aid. I heard Sam Harris on a podcast recently talk about his bullishness on the humanities (paraphrasing: people don't care if a human reads their MRI if detection is good, but people probably do care that a human wrote the novel they're reading).
Music might even be a better example of the irreplaceability of people. While some people might bop along to a tune composed by Suno on the radio, live music is just so much more enjoyable for me. And even better than listening to a live show played by masters, is playing together with friends. To the extent that AI can patiently help us learn the skills to express our own creativity, I'm here for it!
elasticdog 4 hours ago [-]
For just chord analysis, there's "Harte notation", which is meant to be unambiguous representation of the notes (https://ismir2005.ismir.net/proceedings/1080.pdf). That obviously doesn't get you all of the additional information necessary for engraving and full representation of the music, but there are research datasets available using it like https://github.com/smashub/choco. I've also used the https://github.com/MarkGotham/When-in-Rome dataset for some analysis work, but again that's not 100% what you're looking for.
You might like the "iReal Pro" app for the replacement and transposition of jazz standards on your tablet. It's pretty great for that use case versus camera scans.
I forgot to mention ABC. I have seen a few LLMs look at that. There was a model / paper published a couple years back called ChatMusician that built around it.
With the caveat that I'm not terribly fluent in ABC, it seems to me that simple things are simple, but hard things seem to be nearly pathological. And (again, maybe a lapse in my understanding) it seems like there may be a fair number of concepts that are impossible to convey in ABC?
Lastly, if I understand correctly, ABC got its start and is mostly popular as a simplified format for church songbooks. I'd imagine that would, uh, influence the training corpora towards sounding a bit... church songbooky.
EDIT: I may have been overly dismissive of ABC on first glance. It does seem like people have extended it quite a bit, and that it's at least, in theory, capable of encoding most of what I'd expect. And it's human readable, which is a benefit. Though, readability does take a stiff penalty the more richness you add (e.g. dynamics, articulations, stacked notes, etc)
WhitneyLand 5 hours ago [-]
The simplicity is really cool.
To let LLMs compose music I chose json for context efficiency, but this seems like it could be better choice, simple, efficient, already a real format.
Create a benchmark for this problem that researchers can easily run and the problem will solve itself.
mcbetz 6 hours ago [-]
I observe that music OCR space and the only really good solution so far is soundslice. You scan and review some edge cases and get really good results. Paid service by a small company, very worthy to be supported!
peatmoss 5 hours ago [-]
I just signed up a trial, and uploaded a messy Real Book scan. It did very well! It missed the coda markings, but then again the directive in the Real Book was nonstandard. I guess that's a case where a multimodal model might have been able to read the text ("after solos, D.C. al coda") and do something smarter.
WhitneyLand 6 hours ago [-]
“there aren't great corpora of training data that would connect a MusicXML representation to sheet music images or to audio”
It may not be necessary…a lot of the training pairs/data for this could probably be procedurally created via code.
Would be pretty fun to work on and see it come to life.
peatmoss 6 hours ago [-]
I'd imagine that rendered audio that just used midi voices (even high quality "Real Instruments" midi voices) would be pretty brittle for e.g. stem separation or automatic transcription. In a best case, I think you'd start with a clean digital representation, render sheet music imagery, and then have lots of recordings by a bunch of real instrumentalists playing the same music.
On the topic of stem separation, I've wondered about creating a quasi-synthetic dataset by taking chunks of recordings by real musicians playing them back in a real space in various combinations and recording the resulting analog-blended cacophony. Could repeat in various environments like cathedrals, basement bars, etc for realism :-)
...but why is it so complicated? A novice interpretation of "music" is "a bunch of notes!" ... my amateur interpretation of "music" is "layers of notes".
You can either spam 100 notes in a row, or you effectively end up with:
melody = [ a, b, [c+d], e, ... ]
bassline = [ b, _, b, _, ... ]
music = melody + bassline
score = [
"a bunch of helper text",
+ melody,
+ bassline,
+ page_size, etc...
]
...so Lilypond basically made "Tex4Music", and the format serves a few dual purposes...[snip]
aidenn0 5 hours ago [-]
As someone who has never looked at a jazz score, can you share an example of how jazz sheet music would benefit from different fonts?
peatmoss 4 hours ago [-]
It's just an entrenched aesthetic preference. Jazz fonts (fonts in this context refers both to the words and the music symbols) tend to be quite heavy with thick lines. I've heard that the thick hand-written style was originally to make charts more readable in dimly lit clubs, but with tablets and such, that's an anachronism now.
You can look at samples of Hal Leonard's Real Book(s) on their website to get a sense of what it looks like. Again, just an aesthetic preference, but one I and many others hold nonetheless.
elasticdog 4 hours ago [-]
I also don't love the conventional handwritten aesthetic you often see for jazz fonts. For a project I've been working on, I ended up pulling the handful of chord symbol glyphs out of MuseScore's Leland Text font and adjusting them for use in the UI since I couldn't find a suitable option out there.
arboles 6 minutes ago [-]
I'm going to sound like I live under a rock, but what is the true reason companies open-source genuinely good software?
Shouldn't Baidu (or Google) hoard it for themselves to extract the value in a way the competition isn't be able to imitate?
KitN 7 hours ago [-]
"We would like to thank Deepseek-OCR, Deepseek-OCR-2, PaddleOCR for their valuable models and ideas."
Class Act.
gcr 7 hours ago [-]
I don’t understand the shade being thrown ?
nickspacek 7 hours ago [-]
It's the opposite of shade, unless GP is being sarcastic. "Class act" is normally a compliment, and in the context here it sounds to me like they're congratulating Baidu/the researchers in being transparent about where their ideas came from.
pbhjpbhj 6 hours ago [-]
To be fair, I think I see "[real] class act" almost always used sarcastically.
squidbeak 2 hours ago [-]
I've never seen it used that way.
Any compliment can be repurposed as sarcasm, but it's obscenely cynical to immediately assume a compliment is sarcastic - instead of just a compliment. And by the way, there's no 'real' in the poster's message.
pbhjpbhj 19 minutes ago [-]
Yet you responded to my comment in the most cynical way possible. I was excusing the misunderstanding -- I assumed that the parent might only have seen it used cynically as that is a charitable way to interpret the apparent miscommunication and is quite possible. You shit on my comment and then told me doing such things is "cynical". Imagine a pipe before the closing square-bracket if you wish -- the standard editorial convention as I learnt it was that square-bracketed terms may be present or not (in English Language prose).
Cheers.
novoreorx 5 hours ago [-]
FYI, "Unlimited OCR Works" is a Fate/stay night reference. The original "Unlimited Blade Works" is a magic whose entire premise is copying weapons other people forged
aliljet 5 hours ago [-]
How does this compare with infinty parser 2 which seemed to be running the table on every other OCR tool (https://huggingface.co/datasets/allenai/olmOCR-bench). To be fair, there's no single winning OCR benchmark and this isn't showing up anywhere yet..
(As a side note, I do OCR locally as a small RAG for citations I read in books and also chunk input, but merely to save RAM - interesting this natural approach also work in a streaming model)
lacoolj 3 hours ago [-]
This looks more promising than what Mistral just launched (coincidence?????? i think not.)
This approach feels like it could be used for image gen as well (in some combination). Read/view image, start drawing image using illustrator/inkscape/etc (or just SVG), then fill in with what was missed after
pmarreck 7 hours ago [-]
my attempts at using AI to do OCR have always resulted in invented artifacts, which is not production feasible. does this suffer from that as well?
A simple example is words that are supposed to be in other languages being automatically translated to English, which ruins the effect
pbhjpbhj 5 hours ago [-]
You almost don't want [super-]word level ML (ie word-pair/phrase/sentence/document/corpus level).
In transcription, you want near certainty, or you want marking that the word could not be read with certainty - yes, context lets you guess, but you want - for some OCR - to know when it's a guess based on other than the letters in order forming a word.
Example, in a census document on familysearch.com the transcriber "corrected" a name as Joseph. The literal letters in the handwritten document spell Josepth ... and sure enough that's a local variant spelling (Eire).
In another document the writer has used "Joh" as an abbreviation, a [human, I assume] transcriber put that as John ... which is most likely, but happens to be wrong.
Sometimes you care that it's guessed, sometimes you want just the best guess.
messe 4 hours ago [-]
> Eire
A nitpick, because it's often a dogwhistle: but almost nobody in Ireland calls it that when speaking English. And that's still incorrect in Irish, the correct spelling is Éire.
pbhjpbhj 28 minutes ago [-]
By saying it's a dogwhistle are you saying that not adding the correct diacritics is considered racist by Irish people? If I change the rest of the sentence to Na Gaeilge will that be better.
peterderivaz 3 hours ago [-]
I've been trying out this model on a 4090 to transcribe a Japanese grammar pdf (written in English with lots of Japanese examples) and it seems to be working very well from the small parts I have double checked. The output contains both the kanji/hiragana and English as appropriate without attempting any translation.
It has converted about 200 pages in an hour.
drakmo 6 hours ago [-]
If I would want to achieve 100% recognition results I would combine this method with an image model recreating the original document from the transcribed text and matching the layout. One can do that with using all but the page or paragraph from the document you want to recreate (to avoid recreating the exact passage under test from the image artifact directly). After reconstructing you can do an optical comparison that specifically matches misaligned characters and find the errors. Rinse and repeat. Expensive but it would guarantee 100% recognition.
aliljet 4 hours ago [-]
I'm curious about this. What models/tools have you been using?
manipalite 7 hours ago [-]
Whatever happened to Reducto, was very promising 12-15 months ago
gettingoverit 5 hours ago [-]
How does it compare against Finereader? Comparisons against transformer-based OCRs don't really tell anything. The last time I checked, neither of them were of "OCR this legal document" quality.
6 hours ago [-]
overflowy 6 hours ago [-]
What are the requirements for running this locally?
piterrro 5 hours ago [-]
can someone explain how is this different than feeding the VLM model one page at a time?
functional_dev 4 hours ago [-]
[dead]
alansaber 6 hours ago [-]
We've invented chunking? We are so back.
ahknight 6 hours ago [-]
Streaming.
aozelai 5 hours ago [-]
[flagged]
jingpostmedia 4 hours ago [-]
[flagged]
madikz 7 hours ago [-]
[flagged]
swordlucky666 6 hours ago [-]
[dead]
shevy-java 6 hours ago [-]
Is this an academic paper that is published in year xyz,
but in +5 years nobody will remember it anymore?
ramon156 7 hours ago [-]
I love that the entire goal is to push Deepseek OCR further. The west can learn greatly from these companies
Oras 7 hours ago [-]
OCR has been solved long time ago with vision models. Solutions are consistent, reliable, and stable. What is the point of reinventing the wheel?
I would definitely understand post processing, like extracting data, answering question .. etc, but why re-doing the OCR engine itself?
joss82 7 hours ago [-]
I've been working on Parseur for the last 10 years, and OCR has not been solved yet, let me tell you.
OCR still sucks in 2026. Hopefully this might improve the situation but I haven't tested it yet.
chpatrick 7 hours ago [-]
It absolutely hasn't been solved, it's just got pretty decent in recent years.
malfist 6 hours ago [-]
Pretty decent might be quiet the stretch. I'd term it almost acceptable, but only if you're using commercial solutions like amazon's textract, doing it with open source tools is at best, extremely painful and vaguely accurate.
chpatrick 4 hours ago [-]
PaddleOCR (also from Baidu) is pretty damn good actually.
__rito__ 3 hours ago [-]
I have shipped with PaddleOCR to prod. Works pretty well. (Usage limited to printed documents in Anglosphere). Runs fully offline, in CPU.
gettingoverit 5 hours ago [-]
Is it? I've never seen a single OCR that would replace a human just typing it by hand.
What if the goal is something actually useful, such as converting scientific paper PDF back to LaTeX that renders into a pixel-perfect copy? What about converting tables from electronics datasheets into computer-readable form? I wouldn't even expect it in the next decade.
SyneRyder 4 hours ago [-]
I've had success with vision models & OCR, saved me many hours / days / weeks of typing work.
Last year I finally OCR'd many hundreds of pages of my father's old writings. I found that feeding it to Claude Sonnet 4.x via API gave me results that were perfect. No corrections required. So perfect, that Claude was reading along with the story, and actually pointed out a continuity error in the story where an incorrect character was reciting dialog. Claude asked if it should transcribe exactly as is or if I would like Claude to correct the continuity error.
Claude also correctly OCR'd some handwriting that was in the margins of the documents. Sonnet came very close to transcribing a Word Sleuth puzzle, but that was where I hit the limits of its capability at the time.
Mistral OCR was also good (and actually what I started with), but it wasn't quite as good as Claude. And when it was wrong, Mistral could be frighteningly wrong - one API call must have failed, the model must have been presented with a pure black / null image, and I got back a "transcription" that described neverending darkness. It read like something the Woodsman would have broadcast in Twin Peaks S3E8. That poor model.
Tables from electronics datasheets might be okay, I think I've had success with OCR of technical manuals with tables for 80s synthesizer hardware. But I admit my use cases don't crossover into transcriptions of equations or graphs.
sscaryterry 7 hours ago [-]
Detecting characters almost, layout no.
wongarsu 7 hours ago [-]
Exactly my experience. If you try to OCR hand-filled forms with a fixed structure, traditional OCR models are great. Vision-llms can improve a bit on character recognition, but at the cost of harder to detect failure modes.
But if you are trying to ingest diverse documents with headings, multi-column layouts, headers and footers, ad space in the middle of your text, etc, vision-llms are a giant step forward. But you need the context of the previous page to make good decisions about the current page, which is where things quickly get janky (or slow, if you choose the naive approach)
Vision-llms also seem to deal much better with variance in scripts. Cursive, random Japanese in the middle of the text, weird math symbols, handwriting from three centuries ago, all "just works" without you even having to remember that this can happen
ljouhet 7 hours ago [-]
Real question: what tool do you use? (for long/complex documents with tables, code, maths)
- marker (with --force-ocr) gives me the best results
- Mistral OCR (seems really great, but I never managed to get it work)
- Mathpix (tried a long time ago)
- docling (gives me garbage, I must use it wrong)
- Unlimited OCR (will try it)
- ???
Oras 6 hours ago [-]
- Azure Document Intelligence (has an option to return markdown too including headers and footers).
- AWS Textract
badlibrarian 6 hours ago [-]
Exactly. They're both very expensive and prone to surprising you. Sometimes in a good way, sometimes in a bad way. I'd rate them 85%, but you have to run a test because they both fail in different ways on the 15%.
ai_fry_ur_brain 6 hours ago [-]
poma-ai has really great chunking techniques that chunk the document based on the document structure/heirarchy.
We use it on 200 page IEEE standards that are notoriously complex, filled with tables and diagram. Highly reccomend.
vulture916 7 hours ago [-]
I haven't done much long-run OCR, so unsure of the current state, but it would seem they overcome this (from their paper):
"A widely held view is that employing a large language model (LLM) as the decoder allows the model to leverage the prior distribution of language, leading to improved OCR performance. However, the downside is equally evident: as the output sequence lengthens, the accumulated KV cache drives up memory consumption and progressively slows down generation."
mamcx 4 hours ago [-]
Aside: what is the best to read receipts/bank statements/invoices?
cannonpalms 7 hours ago [-]
I guess, in theory, the prior distribution of language would allow for improved performance in some cases, especially where input quality is low.
ta988 7 hours ago [-]
This is already used in OCR, tesseract uses that.
Aboutplants 7 hours ago [-]
lol nope it hasn’t been solved. I deal with this constantly and we still have a longggg ways to go
mschuster91 6 hours ago [-]
> I would definitely understand post processing, like extracting data, answering question .. etc, but why re-doing the OCR engine itself?
Well... the idea seems to be (as far as I understand it, at least) that optical errors and artifacts can now be compensated as the OCR engine is now context-aware.
Say, for example, some random long ass name chemical. It's not going to be in a word correction database, but a context-aware engine (ideally, one that has been supplemented with chemistry data) can now correct "bad" reads of the chemical's name.
Of course, there remains the issue of how to prevent the infamous Xerox bug [1]...
Traditional OCR is faster, cheaper, and much more reliable than LLMs
j16sdiz 7 hours ago [-]
If you consider non-English script, traditional OCR is not more reliable.
CJK have lots of character and high confusion rate.
Arabic scripts are complex and have lots of morphs.
Vietnamese have easily confused diacritics.
Thai have lots of non-standard fonts.
ta988 7 hours ago [-]
I don't think that's a universal statement that aplies to every kind of documents and languages. Mistral OCR is able to do things no "traditional" OCR was ever able to.
JodieBenitez 7 hours ago [-]
I wish it were. Alas...
JohnKemeny 7 hours ago [-]
OCR has definitely not "been solved long time ago", what are you talking about?
The way I understand this works is that the researchers found a clever architectural hack to stop AI from hoarding memory when reading long documents.
Normally, when an AI transcribes a 100 page PDF, it tries to remember every single word it has already ingested. This short-term memory (the KV cache) grows linearly O(N) until the model runs out of VRAM and crashes (or caps it) To avoid this, developers are forced to build janky code that chops PDFs into individual pages, processes them one by one, and glues the text back together.
Unlimited OCR uses Reference Sliding Window Attention (R-SWA) to split the AI's focus into two paths:
Global Reference: The AI keeps full, uncompromised sight of the original document image so it never loses context.
Local Generation: The AI restricts its memory of its own typed text to a tight, moving window (like the last 128 words) and safely forgets the rest.
Will be very interesting for local AI and can’t wait to see what the community builds and extends with it!
You have the overriding context, facts that don't change very often at all. The participants names, their backgrounds etc.
Then you have some very fine grained facts (what they ate for breakfast this morning) which might be useful right now, but are irrelevant outside of a general trend over the longer term.
When trying to reconstruct a conversation you really need to find the right balance without pulling in everything that has ever been discussed.
This definitely is worth further investigation.
There is a whole class of tricky documents. A decent (if you ignore the marketing bias) post about this problem can be found here:
https://getomni.ai/blog/ocr-benchmark
[*] SWAG line height, trial and error to figure out the right amount of overlap given LLM error rates, etc.
Knowing algorithms, data structures and their memory and time complexities is very relevant for SWE. I've had teammates that didn't understand them and everything was fine until when it wasn't (scaling and performance issues).
Or, as I put it to a teammate: "Would you rather review the PR of someone that understands the difference between a set and a list or the PR of someone who doesn't?". This was after we interviewed a candidate with ~15 YoE, on paper, that didn't know the difference.
Agree with this; however knowing how to roll your own BFS/LRU/etc isn't -- in that case I'd rather review the PR of someone who understands how to leverage tested and known implementations than the PR of someone who decided to roll their own.
I got digging into the state of optical music recognition and came away concluding that music is basically a greenfield for AI wherever you look. Optical music recognition is pretty terrible. AI understanding of music theory is terrible (actually looking at music that is; LLMs do okay at text descriptions of theory concepts where you can imagine some online texts making it in).
I think the issue is that we still don't have great digital formats that encode the dots on paper that musicians read. Music notation is pretty rich. Midi doesn't capture all of what's needed for symbolic understanding, because it was mostly made for capturing aspects relevant for playback or performance. MusicXML seems to be the closest for a digital format that encodes the information a musician would want, but there aren't great corpora of training data that would connect a MusicXML representation to sheet music images or to audio. I think that's because MusicXML falls short of encoding enough information to engrave music. Tools like MuseScore need to track a bunch of layout information that isn't encodable in MusicXML. Lilypond format is less verbose that MusicXML and contains a bit more information that is useful to the score creators, but most people don't create sheet music in lilypond. (As an aside, Lilypond bums me out with the state of jazz fonts. I hate looking at "legit" scores in jazz context)
I realize this is mildly off topic, but every time I see people making incremental gains on OCR, which to my mind is pretty good, I am reminded of how abysmal OMR is.
To understand why OMR is so neglected is because most people widely underestimate the difficulty of the task. It has a specific blend of the most extreme shapes combined with an extremely complicated graphical grammar...
My first thought was whether MEI format is being added to MuseScore (the sheet music editor I use these days). It looks like it is: https://music-encoding.org/musescore-doc/
As a somewhat related aside, now that the MuseScore people own Hal Leonard and seem to pushing integration with their cloud subscription service, I wonder if they'll see some of these directions as potentially competing with them. I don't think there's anyone who wouldn't love a transposable clean digital version of their Real Books... and if Hal Leonard is in the business of selling Real Books, I can see where good OMR might be a problem for them. I guess piracy of scanned versions is already rampant, so maybe it's a wash.
AIN'T THAT THE TRUTH.
My girlfriend is studying musicology and she has some physical disabilities that make it difficult for her to write things down sometimes. So I try to help her by writing some AI-powered TTS/OCR/etc. apps here and there. It becomes painfully obvious that music was never considered an important part of any AI training dataset, anywhere.
These days, I'm pleasantly surprised by how well Opus 4.8 understands/explains music theory (as you said). But ask him to transcribe/OCR/OMR some sheet music and he'll confidently give you the MusicXML/Lilypond equivalent of "2 + 2 = horse".
I really hope this ignored area will be swept up with the rest of the rising AI wave, but it's still criminally undervalued.
I always think of the nun character against AI in Mrs Davis:
> "Don't give it a name. No one calls Facebook Doug. No one calls Twitter Mary Lou. No one calls them anything, because no one uses them anymore. They use it, and it's not a person. It's code. - Mrs Davis
I've been thinking about what kind of organization could be self-sustaining and also produce good music AI training data as a natural byproduct. An ideal arrangement would be something that provided some incentive or benefit to musicians in exchange for their recorded interpretation of sheet music. Soundslice, mentioned by another user, seems to do that. They let both teachers and students upload recordings of music that has been turned into MusicXML. The recordings, paired to those snippets of sheet music, has to be a gold mine. Assuming they have enough users. If they aren't already working on stem separation and automatic transcription, they probably should be. Still, my hope would be to figure out some kind of sustainable model where that dataset could be created and released for open model development...
As a domain, I see AI in music as a boon to human creativity. I am very much a novice jazz improvisor, and a passable amateur technician on the trombone. Human instructors can do a lot for me, but there's a lot that is "grinding it out" repetition, where I think AI could be a huge aid. I heard Sam Harris on a podcast recently talk about his bullishness on the humanities (paraphrasing: people don't care if a human reads their MRI if detection is good, but people probably do care that a human wrote the novel they're reading).
Music might even be a better example of the irreplaceability of people. While some people might bop along to a tune composed by Suno on the radio, live music is just so much more enjoyable for me. And even better than listening to a live show played by masters, is playing together with friends. To the extent that AI can patiently help us learn the skills to express our own creativity, I'm here for it!
You might like the "iReal Pro" app for the replacement and transposition of jazz standards on your tablet. It's pretty great for that use case versus camera scans.
With the caveat that I'm not terribly fluent in ABC, it seems to me that simple things are simple, but hard things seem to be nearly pathological. And (again, maybe a lapse in my understanding) it seems like there may be a fair number of concepts that are impossible to convey in ABC?
Lastly, if I understand correctly, ABC got its start and is mostly popular as a simplified format for church songbooks. I'd imagine that would, uh, influence the training corpora towards sounding a bit... church songbooky.
EDIT: I may have been overly dismissive of ABC on first glance. It does seem like people have extended it quite a bit, and that it's at least, in theory, capable of encoding most of what I'd expect. And it's human readable, which is a benefit. Though, readability does take a stiff penalty the more richness you add (e.g. dynamics, articulations, stacked notes, etc)
To let LLMs compose music I chose json for context efficiency, but this seems like it could be better choice, simple, efficient, already a real format.
https://github.com/whitneyland/riffmcp
It may not be necessary…a lot of the training pairs/data for this could probably be procedurally created via code.
Would be pretty fun to work on and see it come to life.
On the topic of stem separation, I've wondered about creating a quasi-synthetic dataset by taking chunks of recordings by real musicians playing them back in a real space in various combinations and recording the resulting analog-blended cacophony. Could repeat in various environments like cathedrals, basement bars, etc for realism :-)
A salient extract:
...but why is it so complicated? A novice interpretation of "music" is "a bunch of notes!" ... my amateur interpretation of "music" is "layers of notes".
You can either spam 100 notes in a row, or you effectively end up with:
...so Lilypond basically made "Tex4Music", and the format serves a few dual purposes...[snip]You can look at samples of Hal Leonard's Real Book(s) on their website to get a sense of what it looks like. Again, just an aesthetic preference, but one I and many others hold nonetheless.
Shouldn't Baidu (or Google) hoard it for themselves to extract the value in a way the competition isn't be able to imitate?
Class Act.
Any compliment can be repurposed as sarcasm, but it's obscenely cynical to immediately assume a compliment is sarcastic - instead of just a compliment. And by the way, there's no 'real' in the poster's message.
Cheers.
(As a side note, I do OCR locally as a small RAG for citations I read in books and also chunk input, but merely to save RAM - interesting this natural approach also work in a streaming model)
This approach feels like it could be used for image gen as well (in some combination). Read/view image, start drawing image using illustrator/inkscape/etc (or just SVG), then fill in with what was missed after
A simple example is words that are supposed to be in other languages being automatically translated to English, which ruins the effect
In transcription, you want near certainty, or you want marking that the word could not be read with certainty - yes, context lets you guess, but you want - for some OCR - to know when it's a guess based on other than the letters in order forming a word.
Example, in a census document on familysearch.com the transcriber "corrected" a name as Joseph. The literal letters in the handwritten document spell Josepth ... and sure enough that's a local variant spelling (Eire).
In another document the writer has used "Joh" as an abbreviation, a [human, I assume] transcriber put that as John ... which is most likely, but happens to be wrong.
Sometimes you care that it's guessed, sometimes you want just the best guess.
A nitpick, because it's often a dogwhistle: but almost nobody in Ireland calls it that when speaking English. And that's still incorrect in Irish, the correct spelling is Éire.
It has converted about 200 pages in an hour.
I would definitely understand post processing, like extracting data, answering question .. etc, but why re-doing the OCR engine itself?
OCR still sucks in 2026. Hopefully this might improve the situation but I haven't tested it yet.
What if the goal is something actually useful, such as converting scientific paper PDF back to LaTeX that renders into a pixel-perfect copy? What about converting tables from electronics datasheets into computer-readable form? I wouldn't even expect it in the next decade.
Last year I finally OCR'd many hundreds of pages of my father's old writings. I found that feeding it to Claude Sonnet 4.x via API gave me results that were perfect. No corrections required. So perfect, that Claude was reading along with the story, and actually pointed out a continuity error in the story where an incorrect character was reciting dialog. Claude asked if it should transcribe exactly as is or if I would like Claude to correct the continuity error.
Claude also correctly OCR'd some handwriting that was in the margins of the documents. Sonnet came very close to transcribing a Word Sleuth puzzle, but that was where I hit the limits of its capability at the time.
Mistral OCR was also good (and actually what I started with), but it wasn't quite as good as Claude. And when it was wrong, Mistral could be frighteningly wrong - one API call must have failed, the model must have been presented with a pure black / null image, and I got back a "transcription" that described neverending darkness. It read like something the Woodsman would have broadcast in Twin Peaks S3E8. That poor model.
Tables from electronics datasheets might be okay, I think I've had success with OCR of technical manuals with tables for 80s synthesizer hardware. But I admit my use cases don't crossover into transcriptions of equations or graphs.
But if you are trying to ingest diverse documents with headings, multi-column layouts, headers and footers, ad space in the middle of your text, etc, vision-llms are a giant step forward. But you need the context of the previous page to make good decisions about the current page, which is where things quickly get janky (or slow, if you choose the naive approach)
Vision-llms also seem to deal much better with variance in scripts. Cursive, random Japanese in the middle of the text, weird math symbols, handwriting from three centuries ago, all "just works" without you even having to remember that this can happen
- marker (with --force-ocr) gives me the best results
- Mistral OCR (seems really great, but I never managed to get it work)
- Mathpix (tried a long time ago)
- docling (gives me garbage, I must use it wrong)
- Unlimited OCR (will try it)
- ???
- AWS Textract
We use it on 200 page IEEE standards that are notoriously complex, filled with tables and diagram. Highly reccomend.
"A widely held view is that employing a large language model (LLM) as the decoder allows the model to leverage the prior distribution of language, leading to improved OCR performance. However, the downside is equally evident: as the output sequence lengthens, the accumulated KV cache drives up memory consumption and progressively slows down generation."
Well... the idea seems to be (as far as I understand it, at least) that optical errors and artifacts can now be compensated as the OCR engine is now context-aware.
Say, for example, some random long ass name chemical. It's not going to be in a word correction database, but a context-aware engine (ideally, one that has been supplemented with chemistry data) can now correct "bad" reads of the chemical's name.
Of course, there remains the issue of how to prevent the infamous Xerox bug [1]...
[1] https://www.dkriesel.com/en/blog/2013/0802_xerox-workcentres...
CJK have lots of character and high confusion rate.
Arabic scripts are complex and have lots of morphs.
Vietnamese have easily confused diacritics.
Thai have lots of non-standard fonts.
In your opinion, what is SOTA here?