Apertus 1.5: From Open Text Models to Open Multimodal Reasoners
Paper • 2610.15542 • Published • 42
AI & ML interests
Apertus is a project of the @swiss-ai initiative. This organization is currently being used for pre-release and testing purposes. Please contact @loleg if you have any questions.
Recent Activity
View all activity
atcbosselut
authored 18
papers 10 days ago
Counterfactual Story Reasoning and Generation
Paper • 1909.04076 • Published
VinaBench: Benchmark for Faithful and Consistent Visual Narratives
Paper • 2503.20871 • Published
ComFact: A Benchmark for Linking Contextual Commonsense Knowledge
Paper • 2210.12678 • Published
A Logical Fallacy-Informed Framework for Argument Generation
Paper • 2408.03618 • Published
Large Language Models Align with the Human Brain during Creative Thinking
Paper • 2604.03480 • Published • 7
CresOWLve: Benchmarking Creative Problem-Solving Over Real-World Knowledge
Paper • 2604.03374 • Published
AI Meets Mathematics Education: A Case Study on Supporting an Instructor in a Large Mathematics Class with Context-Aware AI
Paper • 2603.26679 • Published • 1
Brittlebench: Quantifying LLM robustness via prompt sensitivity
Paper • 2603.13285 • Published
Measuring what Matters: Construct Validity in Large Language Model Benchmarks
Paper • 2511.04703 • Published • 8
COMET-ATOMIC 2020: On Symbolic and Neural Commonsense Knowledge Graphs
Paper • 2010.05953 • Published
Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization
Paper • 2508.04796 • Published • 1
GeoExplorer: Active Geo-localization with Curiosity-Driven Exploration
Paper • 2508.00152 • Published
PERK: Long-Context Reasoning as Parameter-Efficient Test-Time Learning
Paper • 2507.06415 • Published • 9
ConLID: Supervised Contrastive Learning for Low-Resource Language Identification
Paper • 2506.15304 • Published • 1
Mixture of Cognitive Reasoners: Modular Reasoning with Brain-Like Specialization
Paper • 2506.13331 • Published • 2
Context is Gold to find the Gold Passage: Evaluating and Training Contextual Document Embeddings
Paper • 2505.24782 • Published • 1
Creative Preference Optimization
Paper • 2505.14442 • Published
hannayukhymenko
submitted a
paper to Daily Papers 11 days ago
Post
130
Hosted in Lugano in collaboration with SUPSI research staff 📍 SME Circle #5 will be a chance to connect local AI expertise, tour the computing facility, hear the latest news about Apertus and the Swiss AI Initiative at ETHZ EPFL CSCS SwissICT 🇨🇭 Unisciti a noi! https://luma.com/9atdlga7