10 papers
System One models Are Not Better Rerankers
Michael Dinzinger, Kanishka Ghosh Dastidar, Michael Granitzer
System One models, or decision models, answer a question about an input with a calibrated probability instead of generated text. Within weeks of the first release, in September 202…
The WebKurator.de Platform: Combined Regional and Topical Web Curation
Michael Dinzinger, Natanael Arndt, Ben Böck +2
The systematic curation of the Web remains a central challenge for national libraries and memory institutions that aim to preserve culturally and regionally relevant content. Exist…
When Is Complex Chunking Worth It? A Multi-Objective Evaluation of Chunking Methods at Scale
Laura Caspari, Kanishka Ghosh Dastidar, Michael Dinzinger +2
Dense retrieval is commonly evaluated on benchmarks that represent each document with a single embedding, even though real-world retrieval systems often index long documents that r…
Form Without Function: Agent Social Behavior in the Moltbook Network
Saber Zerhoudi, Kanishka Ghosh Dastidar, Felix Klement +9
Moltbook is a social network where every participant is an AI agent. We analyze 1,312,238 posts, 6.7~million comments, and over 120,000 agent profiles across 5,400 communities, col…
WebFAQ 2.0: A Multilingual QA Dataset with Mined Hard Negatives for Dense Retrieval
Michael Dinzinger, Laura Caspari, Ali Salman +3
We introduce WebFAQ 2.0, a new version of the WebFAQ dataset, containing 198 million FAQ-based natural question-answer pairs across 108 languages. Compared to the previous version,…
OwlerLite: Scope- and Freshness-Aware Web Retrieval for LLM Assistants
Saber Zerhoudi, Michael Dinzinger, Michael Granitzer +1
Browser-based language models often use retrieval-augmented generation (RAG) but typically rely on fixed, outdated indices that give users no control over which sources are consult…
Compressed Concatenation of Small Embedding Models
Mohamed Ayoub Ben Ayad, Michael Dinzinger, Kanishka Ghosh Dastidar +2
Embedding models are central to dense retrieval, semantic search, and recommendation systems, but their size often makes them impractical to deploy in resource-constrained environm…
CoRECT: A Framework for Evaluating Embedding Compression Techniques at Scale
L. Caspari, M. Dinzinger, K. Ghosh Dastidar +3
Dense retrieval systems have proven to be effective across various benchmarks, but require substantial memory to store large search indices. Recent advances in embedding compressio…
WebFAQ: A Multilingual Collection of Natural Q&A Datasets for Dense Retrieval
Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar +2
We present WebFAQ, a large-scale collection of open-domain question answering datasets derived from FAQ-style schema.org annotations. In total, the data collection consists of 96 m…
A Survey of Web Content Control for Generative AI
Michael Dinzinger, Florian Heß, Michael Granitzer
The groundbreaking advancements around generative AI have recently caused a wave of concern culminating in a row of lawsuits, including high-profile actions against Stability AI an…