TRAILS
TRAILS
Home
News
People
Publications
Contact
Light
Dark
Automatic
Josef van Genabith
Professor at German Research Center for Artificial Intelligence (DFKI)
DFKI
About Me
Latest
Dfki-mlt at semeval-2026 task 7: Steering multilingual models towards cultural knowledge
Multilingual steering by design: Multilingual sparse autoencoders and principled layer selection
Grounding or Guessing? Visual Signals for Detecting Hallucinations in Sign Language Translation
CLaS-Bench: A Cross-Lingual Alignment and Steering Benchmark
Modular Arithmetic: Language Models Solve Math Digit by Digit
Language Arithmetics: Towards Systematic Language Neuron Identification and Manipulation
Multilingual Political Views of Large Language Models: Identification and Steering
On Multilingual Encoder Language Model Compression for Low-Resource Languages
TenseLoC: Tense Localization and Control in a Multilingual LLM
The Lookahead Limitation: Why Multi-Operand Addition is Hard for LLMs
Automatic Fact-checking in English and Telugu
Cross-Lingual Fact Verification: Analyzing LLMs Performance Patterns Across Languages
When Scale Meets Diversity: Evaluating Language Models on Fine-Grained Multilingual Claim Verification
Reverse Probing: Evaluating Knowledge Transfer via Finetuned Task Embeddings for Coreference Resolution
MMAR: Multilingual and Multimodal Anaphora Resolution in Instructional Videos
A Comparison of Different Tokenization Methods for the Georgian Language
Cite
×