Indic languages

Hindi, Bengali, Tamil, Marathi, Gujarati, Kannada, Telugu, Urdu — where good sentences make a bad document.

--:--:-- [PST]
8 languages
in a 2026 discourse benchmark with human-annotated pronoun and cohesion alignments
IndicDISCO-MT, EAMT 2026
High quality, still failing
models reach high overall translation quality and do not preserve discourse-level phenomena
IndicDISCO-MT, EAMT 2026
344 million
native Hindi speakers — and Hindi is still low-resource in the technical sense
S01

What the evidence says

A 2026 benchmark built specifically to test discourse in Indian-language translation — eight languages, with human-annotated alignments for pronouns and lexical cohesion — reached a conclusion worth quoting: models achieve high overall translation quality and still fail to preserve discourse-level phenomena.

That sentence describes the exact failure that sentence-level scoring cannot see. Every segment passes. The document does not hold together.

S02

Why it happens

Three things at once. The source languages are morphologically rich, so a single word carries information English spreads across several. Syntactic structures diverge sharply from English word order. And pronouns frequently do not mark gender, so resolving who a sentence is about requires inference across sentences — which a system translating segment by segment cannot do, because it never sees them together.

Add the data problem. These are low-resource languages in the technical sense — the measure is how much digitised text exists to train on, not how many people speak them. Hindi has around 344 million native speakers and is low-resource. The international evaluation that added Bhojpuri in 2025 had to fall back to a simpler metric, because the organisers judged the reliability of the standard metrics unestablished for it. In 2026, a dedicated low-resource task pairs Arabic with Hindi, Bengali, Indonesian and Urdu specifically because these pairs continue to face limited availability of parallel training data.

S03

What it means for your workflow

Segment-level quality estimation will pass Indic content that is internally incoherent — a referent that changes gender halfway through, a key term rendered three different ways in three sections. The higher your automation, the more of this you ship.

S04

What we do

We review at document level, not segment level: pronoun and referent tracking across the file, and terminology cohesion end to end. For long documents, that is a different check from the one your pipeline is running, and it is the one that catches what your pipeline is built to miss.

S05

Send us a long Indic document, not a sample of segments

The failure is between the segments. A sample of segments cannot show it to you.

Send us a file to evaluate