diff --git a/gallery/index.yaml b/gallery/index.yaml index 7c4e86b40082..647bc942ff0f 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -3215,6 +3215,28 @@ - filename: UwU-7B-Instruct-Q4_K_M.gguf sha256: 279b2ba20d51bb155c8dd497cf49e0c28407b1822c75de88cfd83d13fd14a59f uri: huggingface://bartowski/UwU-7B-Instruct-GGUF/UwU-7B-Instruct-Q4_K_M.gguf + +- !!merge <<: *qwen25 + name: "drt-o1-14b" + urls: + - https://huggingface.co/Krystalan/DRT-o1-14B + - https://huggingface.co/bartowski/DRT-o1-14B-GGUF + description: | + This repository contains the resources for our paper "DRT-o1: Optimized Deep Reasoning Translation via Long Chain-of-Thought" + In this work, we introduce DRT-o1, an attempt to bring the success of long thought reasoning to neural machine translation (MT). To this end, + + ๐ŸŒŸ We mine English sentences with similes or metaphors from existing literature books, which are suitable for translation via long thought. + ๐ŸŒŸ We propose a designed multi-agent framework with three agents (i.e., a translator, an advisor and an evaluator) to synthesize the MT samples with long thought. There are 22,264 synthesized samples in total. + ๐ŸŒŸ We train DRT-o1-8B, DRT-o1-7B and DRT-o1-14B using Llama-3.1-8B-Instruct, Qwen2.5-7B-Instruct and Qwen2.5-14B-Instruct as backbones. + + Our goal is not to achieve competitive performance with OpenAIโ€™s O1 in neural machine translation (MT). Instead, we explore technical routes to bring the success of long thought to MT. To this end, we introduce DRT-o1, a byproduct of our exploration, and we hope it could facilitate the corresponding research in this direction. + overrides: + parameters: + model: DRT-o1-14B-Q4_K_M.gguf + files: + - filename: DRT-o1-14B-Q4_K_M.gguf + sha256: 9619ca984cf4ce8e4f69bcde831de17b2ce05dd89536e3130608877521e3d328 + uri: huggingface://bartowski/DRT-o1-14B-GGUF/DRT-o1-14B-Q4_K_M.gguf - &smollm ## SmolLM url: "github:mudler/LocalAI/gallery/chatml.yaml@master"