The ATA Chronicle - January/February 2021 - 34

RR
In TMs, one part of the

" MT systems shouldn't just adapt only when the
user corrects mistakes. They should adapt to
project-specific resources upfront. "
Samuel Läubli

sentence might be perfectly

the same thing. Therefore,

isn't matched at all. If you

we're currently adding

then need to insert 10 words,

multiple MT proposals in

typing or maybe dictation
are great, but handwriting
and finger touch input
this setting. However, for

it's going to work well. The

low-quality MT, you might
want to re-translate larger

same holds true for touch

portions of the segment as

reordering, eye tracking, or

well, where again typing

mid-air gestures: stronger

and dictation are probably

interface changes are

better than other modalities.

required. If you just try to

So, I would rather say that

squeeze it into existing tools,

the new modalities show

I believe the user experience

their benefits for highly

will suffer so strongly that
you'll stick to your mouse and
keyboard. However, we just
open-source released MMPE
on Github in the hope that
people will try it out and give
us additional feedback. (Visit:
https://github.com/
NicoHerbig/MMPE) Who
knows? Maybe even some
CAT developers might decide

matching segments from
TM or high-quality MT

love to see.
I don't disagree regarding

to pre-fill the editing box
with the MT suggestion. If
you would do the same with
highly matching segments
from TMs, it should basically

believe offering multiple
high-quality and diverse
outputs might help.
Especially for shorter
sentences, a translation very
similar to what you aim for is
most likely among the
suggestions. For long
sentences, however, mentally
processing multiple

directly post-editing a single

and moving them somewhere

suggestion. At least this is

else. Here, you produce

what I would expect now.

less and supervise the

We'll know more when we

machine more.

Jost

Why are we stuck
with the concept of

post-editing one MT

suggestions simultaneously
by using mechanisms like
auto-suggest (which would
mean that we don't even have
to look at the various MT
suggestions-we just see
what matches our
keystrokes)? Also-and I
think that Samuel already
alluded to this-why don't we

run a study on this.
In parallel, we're also
looking at more interactive
ways to post-edit, where
you click on parts of the MT
suggestion you don't agree
with and get alternatives.
I believe that this, in
combination with touch
input and handwriting, could
really be a nice approach to
post-editing and could also
work well on tablets.

Samuel

Nico brings up
an important

point. Showing too many
alternatives could lead to
cognitive friction. The

look at a closer integration of

prototype of Lilt11 offered both

that the new interaction

our three most important

what you refer to as auto-

possibilities mostly make

resources (TM, MT, and TBs)

suggest-a single suggestion

sense to quickly fix a variety

with each other and achieve

for word, phrase, or sentence

of smaller changes.

better results that way?

completion that adapts to the

be the same. I just believe

34

quicker than that. But we

cognitively demanding than

we could harvest several MT

our study, we also chose

single suggestion would be

quickly grabbing a few words

modalities might also help

can only guess here. For

a normal post-edit of a

longer and be more

for instance, looking at how

haven't tested that, so I

suggestion three times, since

remaining mistakes, like

TM. The newly explored

scores are high. We just

wants to see almost the same

suggestions might just take

suggestion? Why are we not,

especially when the match

similar MT outputs. No one

very quickly change the few

your comments about

with TM-based work,

MMPE, where we penalize

because they allow you to

to build certain aspects into
their tools, which I would

Indeed, we've been
asking ourselves

matched while another part

might be less helpful in
editing, there's no way that

Nico

The ATA Chronicle | Januar y/February 2021

www.atanet.org
https://www.github.com/ http://www.atanet.org
The ATA Chronicle - January/February 2021

Table of Contents for the Digital Edition of The ATA Chronicle - January/February 2021
