Adam Bittlingmayer (@bittlingmayer) 's Twitter Profile
Adam Bittlingmayer

@bittlingmayer

translation quality prediction @modelfront Inschenör, Dolmetscher, πολυί̈στωρ, libertaire, օտար, rom po drom, 𒀲... Not available in your country.

ID: 26206117

linkhttps://linkedin.com/in/bittlingmayer calendar_today24-03-2009 09:31:42

1,1K Tweet

689 Followers

1,1K Following

ModelFront (@modelfront) 's Twitter Profile Photo

How to accelerate translation? 📈 Join the panel on AI that scores AI - now available in a TMS near you. Featuring: Alon Lavie , VP of AI Research at Hacked Phrase PoC Amir Kamran , Solution Architect at TAUS Adam Bittlingmayer , CEO at ModelFront lu.ma/accelerating-t…

ModelFront (@modelfront) 's Twitter Profile Photo

DLD AI Summit is bringing together AI founders, leaders and investors from ModelFront, Aleph Alpha, IBM, Google, Palantir, Google DeepMind, Microsoft, SAP, Hubert Burda Media, Burda Principal Investments, 468 Capital and more! Wednesday Munich 🇩🇪 dld-conference.com/conference/AIs…

ModelFront (@modelfront) 's Twitter Profile Photo

“Full self-driving" translation is now a reality, with @ModelFront launching AI that controls when to stop and let a human take over...

LT-Innovate (@ltinnovate) 's Twitter Profile Photo

Join us @ Language Intelligence / Semantics 2023 #LI2023 SEMANTiCS Conference Leipzig 21-22/09/23: language-intelligence.com The most convenient way to update yourself on the latest developments in #multilingual#AI: #translation, #speech & #analytics in #machinetranslation

Join us @ Language Intelligence / Semantics 2023 #LI2023 <a href="/SemanticsConf/">SEMANTiCS Conference</a> Leipzig 21-22/09/23: language-intelligence.com The  most convenient way to update yourself on the latest developments in #multilingual#AI: #translation, #speech &amp; #analytics in #machinetranslation
Rico Sennrich (@ricosennrich) 's Twitter Profile Photo

A simple strategy to mitigate hallucinations and off-target translations in machine translation: in addition to maximizing the probability given the true input, minimize the probability given some contrastive input. 1/4

A simple strategy to mitigate hallucinations and off-target translations in machine translation: 

in addition to maximizing the probability given the true input, minimize the probability given some contrastive input.

1/4
ModelFront (@modelfront) 's Twitter Profile Photo

Fast and efficient translation? 📈🌍 Learn about accelerating translation with AI that scores AI! lu.ma/accelerating-t…

ModelFront (@modelfront) 's Twitter Profile Photo

AI for more efficient post-editing? 📈 2023's deepest event on machine translation quality prediction will feature hands-on builders, users and integrators demoing quality prediction in memoQ and Blackbird.io 🗓️Tomorrow Salt Lake City 🇺🇸 taus.net/events/confere…

AI for more efficient post-editing? 📈

2023's deepest event on machine translation quality prediction will feature hands-on builders, users and integrators demoing quality prediction in <a href="/memoQ_Official/">memoQ</a>  and <a href="/blackbirdio_hq/">Blackbird.io</a>

🗓️Tomorrow

Salt Lake City 🇺🇸

taus.net/events/confere…
Slator (@slatornews) 's Twitter Profile Photo

slator.com/automating-mac… Interest in automating machine #translation quality estimation / prediction 💡increases thanks to the potential for productivity gains 📈

ModelFront (@modelfront) 's Twitter Profile Photo

🦾💪 The "hybrid" paradigm - using AI/automation to control when/where human intelligence is required - is key to systems we rely on. How does it apply to text generation?

🦾💪 The "hybrid" paradigm - using AI/automation to control when/where human intelligence is required - is key to systems we rely on.

How does it apply to text generation?
The Machine Translate Foundation (@machtranslate) 's Twitter Profile Photo

📢 Happening tomorrow, 22 hours from now! Online machine translation meetup Adaptive machine translation Feb 8, 3pm CET / 9am ET machinetranslate.org/meetup

ModelFront (@modelfront) 's Twitter Profile Photo

What are the levels of translation efficiency?  What are the prerequisites for accelerating human translation with AI? Join the Bay Area community at Adobe HQ for the next IMUG meetup to hear from @ModelFront co-founder and CEO Adam Bittlingmayer Wednesday at 6:30

What are the levels of translation efficiency?  What are the prerequisites for accelerating human translation with AI?

Join the Bay Area community at <a href="/Adobe/">Adobe</a> HQ for the next <a href="/i18n_mug/">IMUG</a>  meetup to hear from @ModelFront co-founder and CEO Adam Bittlingmayer

Wednesday at 6:30
ModelFront (@modelfront) 's Twitter Profile Photo

How is AI accelerating translation for the leading global luxury marketplace? 📈 At XTM Live, Alex Katsambas will share the story of how Farfetch rolled out AI to accelerate human-quality translation by ██%. April 11th in NY 🗽

Liling Tan (@alvations) 's Twitter Profile Photo

This is an interesting job post for "model teacher" (modern day data annotator), kinda tell you what's inside the training data. #nlproc 1. The model needs to learn your annotation and makes the user feel like it's magic -> MTurk effect >>

This is an interesting job post for "model teacher" (modern day data annotator), kinda tell you what's inside the training data. #nlproc

1. The model needs to learn your annotation and makes the user feel like it's magic -&gt; MTurk effect

&gt;&gt;
Adam Bittlingmayer (@bittlingmayer) 's Twitter Profile Photo

One of unexpected effects of launching machine translation quality prediction: Now every improvement in your machine translation has a direct impact. Before, with fully manual post-editing, a 10% quality improvement had no direct impact.

IMUG (@i18n_mug) 's Twitter Profile Photo

Video! Last month's IMUG talk by Adam Bittlingmayer on the evolution of translation efficiency. Many thanks to MC tex -Tweeto, ergo sum- and to hosts Xiang Zhao and Adobe #G11n! youtu.be/ZISYtTVoa_Y #IMUG408 #nplroc #L10n #t9n

Andrew Ng (@andrewyng) 's Twitter Profile Photo

I think AI agentic machine translation has huge potential for improving over traditional neural machine translation, and am releasing as open-source a demonstration I'd been playing with as a fun weekend project. Using an agentic workflow, this demonstration (i) Prompts an LLM