Papers by Simon Razniewski
Do Children Texts Hold The Key To Commonsense Knowledge? (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to compiling commonsense knowledge (CSK) struggle with reporting bias, i.e., frequency in text sources is not a good proxy for relevance or truth. |
| Approach: | They propose that children's texts make fewer assumptions on the reader's knowledge and therefore spell out commonsense more explicitly. |
| Outcome: | The proposed approach can be leveraged in language-model-based commonsense knowledge extraction settings, where task-unspecific fine-tuning on small amounts of children texts yields significant improvements. |
QUITE: Quantifying Uncertainty in Natural Language Text in Bayesian Reasoning Scenarios (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing probabilistic reasoning datasets require the model to only rank textual alternatives or use limited set of templates. |
| Approach: | They propose a question-answering dataset that uses probabilistic rules to express degrees of certainty. |
| Outcome: | The proposed model outperforms existing models on all reasoning types . it is available on Github and is expected to be used in clinical documentation . |
PAP2PAT: Benchmarking Outline-Guided Long-Text Patent Generation with Patent-Paper Pairs (2025.findings-acl)
Copied to clipboard
| Challenge: | In patents, the description constitutes more than 90% of the document on average, yet its automatic generation remains understudied. |
| Approach: | They propose a method to generate patent documents using a research paper as an invention specification. |
| Outcome: | The proposed model can generate 1.8k patent-paper pairs describing the same inventions, but it's difficult to provide the level of detail required. |
Evaluating the Knowledge Base Completion Potential of GPT (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Language models (LMs) have been proposed for unsupervised knowledge base completion (KBC) however, their ability to do this at scale and with high accuracy remains an open question. |
| Approach: | They propose to use language models to complete a large public KB, Wikidata, with 90% precision. |
| Outcome: | The proposed models can extend Wikidata by 27M facts at 90% precision. |
ENTYFI: A System for Fine-grained Entity Typing in Fictional Texts (2020.emnlp-demos)
Copied to clipboard
| Challenge: | ENTYFI is a web-based system for fine-grained typing of entity mentions in fictional texts. |
| Approach: | They propose a web-based system for fine-grained typing of entity mentions in fictional texts . entity types are a core building block of current knowledge bases . |
| Outcome: | The proposed system builds on 205 automatically induced high-quality type systems for popular fictional domains and provides recommendations towards reference type systems. |
Foundations of LLM Knowledge Materialization: Termination, Reproducibility, Robustness (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large Language Models encode substantial factual knowledge, yet measuring and systematizing it remains challenging. |
| Approach: | They systematically analyze LLM knowledge materialization using miniGPTKBs . they find high termination rates, though model-dependent, and mixed reproducibility . |
| Outcome: | The proposed model can reliably surface core knowledge, but it has limitations. |
Enabling LLM Knowledge Analysis via Extensive Materialization (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have majorly advanced NLP and AI, and a major success factor is their internalized factual knowledge. |
| Approach: | They propose a method to comprehensively materialize an LLM’s factual knowledge through recursive querying and result consolidation. |
| Outcome: | The proposed method provides constructive insights into the scope and structure of LLM knowledge (or beliefs) it provides scale, accuracy, bias, cutoff and consistency at the same time. |
SANDI: Story-and-Images Alignment (2021.eacl-main)
Copied to clipboard
| Challenge: | a method for selecting images from an image collection and aligning them with text paragraphs of a story is presented . judiciously placed images are used for multimodal descriptions and narration in stories . |
| Approach: | They propose a method for automatically selecting images from an image collection and aligning them with text paragraphs of a story. |
| Outcome: | The proposed method can select and align images with texts with high quality of semantic fit. |
Inside ASCENT: Exploring a Deep Commonsense Knowledge Base and its Usage in Question Answering (2021.acl-demo)
Copied to clipboard
| Challenge: | Structured knowledge bases (KBs) are a great way to explain and interpret outputs of systems leveraging the resources. |
| Approach: | They propose a web portal that allows users to understand its construction process and explore its content. |
| Outcome: | The proposed framework allows users to understand its construction process, explore its content, and observe its impact in the use case of question answering. |
Coverage of Information Extraction from Sentences and Paragraphs (D19-1)
Copied to clipboard
| Challenge: | Textual information extraction (IE) uses textual features to negate stronger statements, such as the negation of stronger statements. |
| Approach: | They propose to use textual features to predict whether a given text segment mentions all objects standing in a certain relationship with a subject. |
| Outcome: | The proposed features can predict whether a given text segment mentions all objects standing in a certain relationship with a particular subject. |
Predicting Document Coverage for Relation Extraction (2022.tacl-1)
Copied to clipboard
| Challenge: | Existing methods for predicting document coverage for relation extraction (RE) are limited in their predictive power. |
| Approach: | They propose a task of predicting the coverage of a text document for relation extraction . they analyze a dataset of 31,366 diverse documents for 520 entities . |
| Outcome: | The proposed model achieves an F1 score of up to 46% on two use cases. |