SenseNova-SI is a series of spatial intelligence models built on multimodal foundation models, specifically optimized for spatial understanding capabilities. Through large-scale spatial intelligence data training, it performs excellently in metric estimation, spatial relationship understanding, viewpoint change handling, etc.

MultimodalTransformers

A semantic embedding model trained on publicly available data from the EU Horizon 2020 IMPETUS Climate project, focusing on climate change, urban resilience, and adaptation strategies.

Natural Language ProcessingSafetensorsEnglish

A text classification model based on XLM-roberta for detecting the salience of climate change issues in political texts

Natural Language ProcessingTransformers

This is an audio segmentation model capable of detecting speaker changes, voice activity, and overlapping speech, suitable for audio analysis in multi-speaker scenarios.

Audio ProcessingPytorch

This is a base model fine-tuned for meteorological and climate data downscaling tasks, built on the Prithvi WxC base model. The model can downscale low-resolution meteorological and climate data to higher resolutions and has been successfully applied to the processing of MERRA-2 data, data from Environment and Climate Change Canada, and EURO-CORDEX climate simulation data.

Scientific ComputingPytorch

This is a text classification model fine-tuned based on XLM-R Large, specifically designed to identify German false information narratives. The model is trained to detect common false information patterns, such as narratives about immigration crimes, climate change denial, election manipulation, etc., and provides support for automatic fact-checking.

Natural Language ProcessingTransformersGerman

This is a speaker segmentation model based on pyannote.audio, capable of detecting speech activity, speaker changes, and overlapping speech.

Audio ProcessingPytorch

The gpt2-chatbot is a dialogue model obtained through supervised fine-tuning (SFT) on the Deita dataset based on the GPT2-XL architecture, aiming to change a certain decision tendency. This model supports multi-turn dialogue and performs well in text generation tasks, but has limitations in mathematical reasoning.

Natural Language ProcessingTransformers

AdaptFormer is an adaptive hierarchical semantic method for remote sensing image change detection, employing differentiated strategies for varying semantic depths.

Computer VisionTransformers

The speaker diarization pipeline of Pyannote.audio, used to automatically detect speaker changes in audio and segment speech segments.

Audio ProcessingPyannote.audio

RobBERT-2023 is the latest language model based on Dutch. It adopts the RoBERTa architecture and uses an updated Dutch tokenizer and training data to adapt to the latest development and changes in Dutch.

Natural Language ProcessingTransformersOther

ClimateGPT is a 7-billion-parameter AI model specifically designed to integrate interdisciplinary research on climate change, developed based on the Llama-2 framework with a focus on climate science.

Natural Language ProcessingTransformersEnglish

ClimateGPT is a 70-billion-parameter large language model specialized in climate science, based on continued pre-training and instruction fine-tuning of Llama-2, focusing on interdisciplinary research integration in climate change

Natural Language ProcessingTransformersEnglish

Speaker diarization pipeline trained on pyannote.audio 3.0.0, supporting automatic voice activity detection, speaker change detection and overlapping speech detection

Audio ProcessingPyannote.audio

A political text summarization model fine-tuned for climate change, environment, and energy issues, based on 7,000 press releases from 66 political parties across 12 countries

Natural Language ProcessingTransformersMultiple Languages

A code difference generation model released by CarperAI, specifically designed for handling text editing operations, capable of generating code change suggestions based on code snippets and natural language descriptions.

Natural Language ProcessingTransformersMultiple Languages

Pythia is a suite of large language models developed by EleutherAI specifically for interpretability research. Pythia-160M is a model version with 160M parameters, trained on the Pile dataset, and provides 143 training checkpoints to facilitate the study of changes in the language model during the training process.

Natural Language ProcessingTransformersEnglish

An ESG text classification model fine-tuned based on FinBERT, which can classify text into nine fine-grained ESG topics, including categories such as climate change, natural capital, pollution and waste, and is specifically used for the analysis of enterprise ESG reports and annual reports.

Natural Language ProcessingTransformersMultiple Languages

Speaker diarization model based on pyannote.audio 2.0 for automatic detection of speaker changes and speech activity in audio

Audio ProcessingPyannote.audio

CodeReviewer is a pre-trained model using code changes and code review data, designed to support code review tasks.

Natural Language ProcessingTransformersOther