On 07.04.2022 at 11 a.m., the 9th lecture of the Living Lab lecture series took place. Klaudia-Doris Thellmann and Bernhard Stadler from TU Dresden talked about Language Models for Code Generation. 

Language Models for Code Generation

Efforts to automatically generate source code from natural language instructions to overcome the human-machine language barrier have existed since the early days of computer science. As a result, numerous approaches have emerged over the last decades, ranging from statistical methods with focus on rule induction or probabilistic grammars to artificial neural networks. In the last few years, large-scale pre-trained neural language models have shown impressive performance on a variety of natural language processing tasks and have also become the basis for a growing number of programming-related tasks such as code completion or synthesizing code from natural language descriptions. 

In this talk, ScaDS.AI scientific researchers Klaudia-Doris Thellmann and Bernhard Stadler gave an overview of state-of-the-art language models for program synthesis. Furthermore, they presented basic characteristics of these models and discussed several of their limitations. A critical weakness of these language models is systematic reasoning, which is crucial for understanding the programming task at hand and generating program code. One possible direction of research that could help alleviate these limitations is the inclusion of structural knowledge – an attempt they have made in this regard and briefly introduced.

You can rewatch this lecture on YouTube.

Living Lab Lecture Series

The Living Lab Lecture Series gives you an in-depth insight into the many research topics of ScaDS.AI Dresden/Leipzig. From Natural Language Processing to Ethics and Moral Code in AI, a great variety of topics are discussed. You can join our lectures every first thursday of the month or watch them on YouTube afterwards. If you have ideas for topics to discuss in the future, please let our Living Lab team know. We suggest for you to regularly check our event calendar, to never miss out on upcoming lectures or other interesting events organized by or in cooperation with our center.

FAQ

How can I participate in the lecture?

You can reach the permanent room for all lectures here: https://tud.link/i8zf

When will the room become accessible to the audience?

The room will be accessible 5 minutes before the start of the lecture.

Who can participate in the Living Lab activities?

The participation is free for everyone.

Do I need to be familiar with topics such as Big Data and AI to benefit from the lectures?

No! Not at all. One of our goals in the Living Lab lecture series is to familiarize everyone with these topics.

Is there any technical requirements to participate in the lecture?

You just need an up-to-date browser such as Firefox, Google Chrome or Chromium. We would also recommend using headphones for better audio quality.

Should I also turn on my camera/microphone during the lecture?

Not unless you would like to! In general, there is no need to have a camera or microphone to participate in the lecture.

In that case, how can I ask a question and provide feedback?

Alongside joining the discussion with your camera and microphone, there is also the possibility to submit your question and comments as a written comment in the Chat section.

How can I easily get informed about future activities of the Living Lab?

• • Follow us on X: @Sca_DS

• • Follow us on LinkedIn: ScaDS.AI Dresden/Leipzig

• • Subscribe to our YouTube channel: https://www.youtube.com/channel/UCv7EDPwovn6T1iaMbhlpFjg

• • Subscribe to our mailing list: https://mailman.zih.tu-dresden.de/groups/listinfo/livinglab

Can I participate with my smartphone?