GeNeSy 2024: Generative Neuro-Symbolic AI Workshop at ESWC 2024 Crete, Greece, May 26-30, 2024 |
Conference website | https://sites.google.com/view/genesy2024/ |
Submission link | https://easychair.org/conferences/?conf=genesy2024 |
Submission deadline | March 14, 2024 |
Generative AI is revolutionising our society by enabling innovative applications in art, language, healthcare, and more, fundamentally transforming the way we create, communicate, and solve complex problems. This sentence was indeed written by a large language model (LLM) – currently one of the most popular neural approaches that is trained to produce outputs through a sequence modelling objective. LLMs are trained to generate outputs that are statistically plausible, realistic, but not necessarily correct. As such, while the language models can generate fluent text and mimic humans on many tasks, trusting their reasoning is hindered by challenges such as coherence, consistency, and explainability.
Meanwhile, Symbolic AI provides sound and well-understood formal reasoning and explanation via knowledge representation that can be inspected to interpret how decisions are made from data, while lacking flexible generalisation to novel inputs. Neuro-Symbolic AI aims to build rich computational AI models, systems and applications by combining neural and symbolic learning and reasoning. It hopes to create synergies among the strengths of neural and symbolic AI while overcoming their complementary weaknesses.
The GeNeSy workshop aims at gathering researchers in Generative and Neuro-Symbolic AI to combine expertise, perspectives, and pioneering works to lay down the foundations for novel methods and paradigms for Generative Neuro-Symbolic AI. GeNeSy will feature novel and already published papers on NeSy methods for reasoning and explanations in multiple modalities, benchmarks and evaluation methods, challenges like commonsense reasoning and human-AI teaming, and reflection on ethical and social implications of GenAI. To the best of our knowledge, GeNeSy is the first workshop that brings ideas from NeSy and GenAI together, enabling discussions between these two communities and facilitating impact on the future development of AI.
Submission Guidelines
Submissions can fall into one of the following categories :
- Full Research Papers (12-15 pages) - Papers with original research work which will be judged on their technical soundness and rigour, though allowances made for novel or experimental directions.
- Short Papers (6-10 pages) - Position papers or reports of new research directions, especially where the work is less mature but nonetheless technically sound.
- Review Papers (8-15 pages) - Articles presenting review of GeNeSy architectures and/or applications.
- Dissemination Papers (1 page abstract) - Already published papers from top AI and semantic web venues such as NeurIPS, WebConf, AAAI, ICML, ICLR, ACL, EMNLP, ISWC that are relevant to the workshop. An abstract explaining the work and its relevance to the workshop is required.
The page limits indicated above are inclusive of references (except dissemination papers which only require an abstract).
Papers in the research, short and review categories will be peer reviewed by at least 2 reviewers. For the long and short papers, we also welcome submissions reporting negative results and sharing experimental insights on the technical challenges and issues in the application of GeNeSy methods.
Dissemination papers will go through short review from the organisers, checking for their quality and relevance to the workshop. They will not be included in the workshop proceedings but presented in the poster session.
Selected papers will be invited to submit to the Special Issue on “Knowledge Graphs and Neurosymbolic AI” of the Neurosymbolic Artificial Intelligence journal.
List of Topics
-
Neuro-Symbolic (NeSy) approaches for data generation, including (but not limited to) text; images and videos; audio; time-series; and multimodal applications
-
Methods for knowledge graph completion and knowledge-augmented explanation
-
Neuro-symbolic methods for knowledge-augmented reasoning
-
Neuro-symbolic methods for generative commonsense reasoning
-
NeSy methods for data quality assessment and evaluation
-
Review of Generative NeSy architectures and tasks
-
Human-centric and cognitive Generative NeSy architectures
-
Trustworthy methods for computational creativity in art and science
-
Frameworks for the validation, verification, and adaptation of Generative AI outputs
-
Applications and expected challenges for Generative NeSy methods
-
Ethical, societal implications and case studies of Generative AI methods
Contact
All questions about submissions should be emailed to genesyworkshop2024@googlegroups.com