BavGround: A Benchmark for Regional Cultural Grounding and Dialect Competence in Bavarian Paper β’ 2608.12894 β’ Published 2 days ago β’ 1
view article Article Meta is back with Muse Glimmer: local, agentic, multimodal, and open source +2 pcuenq, merve, burtenshaw, ariG23498 β’ 5 days ago β’ 91
view article Article FineBooks: are open OCR models good enough to unlock historical knowledge? finebooks β’ 4 days ago β’ 24
view article Article Making Knowledge Distillation Cheap Enough to Run at Scale MultiverseComputingCAI β’ 4 days ago β’ 29
Beyond Sequence Order: Syntax-Informed Positional Embeddings for Transformers Paper β’ 2608.06111 β’ Published 9 days ago β’ 6
When Attention Goes Blind: Numerical Failure in ALiBi Positional Encodings Paper β’ 2608.03994 β’ Published 11 days ago β’ 8
Memory Decoder at Scale: A Pretrained, Parametric Long-Term Memory Paper β’ 2607.27919 β’ Published 16 days ago β’ 58
Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors Paper β’ 2607.25335 β’ Published 18 days ago β’ 1
From Data to Device: ELMOD An Efficient German-First 2.7B Language Model for Mobile Inference Paper β’ 2607.24585 β’ Published 19 days ago β’ 1
ELMOD-2.7b Collection Efficient Language Model for On-Device Deployment (ELMOD) a compact (2.7B) German LM designed for efficient inference on constrained hardware. β’ 2 items β’ Updated 4 days ago β’ 2
Building a European Multilingual Evaluation Dataset: The MMLU Localisation Project within the EMT Network Paper β’ 2607.18432 β’ Published 26 days ago β’ 2
Separating Representation from Reconstruction Enables Scalable Text Encoders Paper β’ 2607.04011 β’ Published Jul 4 β’ 1
Apertus v1.5 Collection 8B and 70B open-data open-weights models, largely multilingual with supported multimodality and reasoning β’ 2 items β’ Updated 10 days ago β’ 31
news-crawler-LM: A Small Long-Context Model For High-Quality News Crawling Paper β’ 2607.21284 β’ Published 23 days ago β’ 2
view article Article Be Ready Before the Attack: A Practical Guide to Self-Hosting an Open Model for Cyber Defense jeffboudier β’ 25 days ago β’ 20
A Sovereign, Open-Source Foundation Model for German and English Paper β’ 2607.09424 β’ Published Jul 10 β’ 15
view article Article Welcome Inkling by Thinking Machines +3 burtenshaw, merve, pcuenq, ariG23498, andito β’ about 1 month ago β’ 164