LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters! Paper β’ 2502.07374 β’ Published 5 days ago β’ 27
Hibiki fr-en Collection Hibiki is a model for streaming speech translation , which can run on device! See https://github.com/kyutai-labs/hibiki. β’ 5 items β’ Updated 10 days ago β’ 48
SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training Paper β’ 2501.17161 β’ Published 19 days ago β’ 105
Reasoning Datasets Collection Distilled synthetic Reasoning datasets β’ 7 items β’ Updated 14 days ago β’ 51
Eagle 2 Collection Eagle 2 is a family of frontier vision-language models with vision-centric design. The model supports 4K HD input, long-context video, and grounding. β’ 9 items β’ Updated 24 days ago β’ 31
view article Article Train 400x faster Static Embedding Models with Sentence Transformers Jan 15 β’ 144
Deepthink and Reasoning Collection Best for Deepthink and Reasoning β’ 14 items β’ Updated 23 days ago β’ 16
ModernBERT Collection Bringing BERT into modernity via both architecture changes and scaling β’ 3 items β’ Updated Dec 19, 2024 β’ 134
Bamba Collection Collection of Bamba - hybrid Mamba2 model architecture based models trained on open data β’ 8 items β’ Updated Dec 18, 2024 β’ 17
Falcon3 Collection Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B parameters. β’ 40 items β’ Updated 3 days ago β’ 80