mlocalllama·LocalLLaMAbymorrowind Dolphin: A Large-Scale Automatic Speech Recognition Model for Eastern Languageshttps://arxiv.org/abs/2503.20212Open linkView original on lemm.ee14Comments2Reply
mlocalllama·LocalLLaMAbymorrowind Sentence transformers v4Link to bluesky https://bsky.app/profile/tomaarsen.com/post/3llc2jvwah22f Some more details https://huggingface.co/blog/train-rerankerView original on lemm.ee28Comments3Reply
mlocalllama·LocalLLaMAbymorrowind NotaGen: Advancing Musicality in Symbolic Music Generation with Large Language Model Training Paradigmshttps://electricalexis.github.io/notagen-demo/Open linkView original on lemm.ee12CommentsReply
mlocalllama·LocalLLaMAbymorrowind StarVector - a foundation model for generating svgshttps://huggingface.co/starvector/starvector-1b-im2svgOpen linkView original on lemm.ee20Comments6Reply
mlocalllama·LocalLLaMAbymorrowind EXAONE Deep ━ Setting a New Standard for Reasoning AI - LG AI Research Newshttps://www.lgresearch.ai/news/view?seq=543Open linkView original on lemm.ee4Comments4Reply
mlocalllama·LocalLLaMAbymorrowind Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketchinghttps://arxiv.org/abs/2503.05179Open linkView original on lemm.ee12Comments1Reply
mlocalllama·LocalLLaMAbymorrowind Sorting-Free GPU Kernels for LLM Samplinghttps://flashinfer.ai/2025/03/10/sampling.htmlOpen linkView original on lemm.ee5CommentsReply
mlocalllama·LocalLLaMAbymorrowind Reka Flash, open source 21B model comparable to QWQ 32BView original on lemm.ee19Comments2Reply
mlocalllama·LocalLLaMAbymorrowind Chain of Draft: Thinking Faster by Writing Lesshttps://arxiv.org/abs/2502.18600Open linkView original on lemm.ee9Comments1Reply
mlocalllama·LocalLLaMAbymorrowind Atom of Thoughts (AOT): lifts gpt-4o-mini to 80.6% F1 on HotpotQA, surpassing o3-mini and DeepSeek-R1https://bsky.app/profile/sungkim.bsky.social/post/3ljgwfe3flk2hOpen linkView original on lemm.ee11CommentsReply
