Forwarded from Анализ данных (Data analysis)
📒 GigaChat нового поколения.
Разработчики @gigachat_bot изменили подход к обучению модели, а потому практически все умения модели были улучшены. В частности, речь идет о сокращении текстов, ответов на вопросы и генерации идей.
Появился и бот GigaChat в социальной сети «ВКонтакте» — после активации, его можно использовать для самых разных целей: от создания текстов до генерации изображений (за счет интеграции с Kandinsky).
Число уникальных пользователей GigaChat достигло 1 млн.
• Попробовать
@data_analysis_ml
Разработчики @gigachat_bot изменили подход к обучению модели, а потому практически все умения модели были улучшены. В частности, речь идет о сокращении текстов, ответов на вопросы и генерации идей.
Появился и бот GigaChat в социальной сети «ВКонтакте» — после активации, его можно использовать для самых разных целей: от создания текстов до генерации изображений (за счет интеграции с Kandinsky).
Число уникальных пользователей GigaChat достигло 1 млн.
• Попробовать
@data_analysis_ml
👍6
PHA
🖥 Github: https://github.com/bumble666/pha
📕 Paper: https://arxiv.org/pdf/2310.11670v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/glue
@ArtificialIntelligencedl
cd PHA
pip install -r requirements.txt🔥 Datasets: https://paperswithcode.com/dataset/glue
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍6❤2🔥2👎1
Zero-shot Faithfulness Evaluation for Text Summarization with Foundation Language Model
🖥 Github: https://github.com/jiaqisjtu/faitheval-fflm
📕 Paper: https://arxiv.org/pdf/2310.11648v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/cnn-daily-mail-1
@ArtificialIntelligencedl
CUDA_VISIBLE_DEVICES=0 python3 main.py🔥 Datasets: https://paperswithcode.com/dataset/cnn-daily-mail-1
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤5👍3🥰1
DD-Net
🖥 Github: https://github.com/fandulu/DD-Net
📕 Paper: https://arxiv.org/pdf/1907.09658.pdf
🔥 Datasets: https://paperswithcode.com/dataset/gtea
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/gtea
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍8❤3🥰1
SplitGNN
🖥 Github: https://github.com/blackboxo/SplitGNN
📕 Paper: https://dl.acm.org/doi/pdf/10.1145/3583780.3615067
🔥 Datasets: https://paperswithcode.com/dataset/fdcompcn
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/fdcompcn
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤9👍3🔥2
SALMONN: Speech Audio Language Music Open Neural Network
🖥 Github: https://github.com/bytedance/salmonn
📕 Paper: https://arxiv.org/pdf/2310.13289v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/librispeech
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/librispeech
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7🔥4❤2
[NeurIPS 2023] Global Structure-Aware Diffusion Process for Low-Light Image Enhancement
🖥 Github: https://github.com/jinnh/GSAD
📕 Paper: https://arxiv.org/pdf/2310.17577.pdf
🔥 Datasets: https://paperswithcode.com/dataset/lol
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/lol
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤10👍4🔥2
Human-Guided Complexity-Controlled Abstractions
🖥 Github: https://github.com/mycal-tucker/human-guided-abstractions
📕 Paper: https://arxiv.org/pdf/2310.17550v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/fashion-mnist
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/fashion-mnist
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤6🔥4🥰1
DialogLLMScenic
🖥 Github: https://github.com/avmb/dialogllmscenic
📕 Paper: https://arxiv.org/pdf/2310.17372v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/carla
⭐ Tasks: https://paperswithcode.com/task/self-driving-cars
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/carla
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7❤4🔥4
PETA: Evaluating the Impact of Protein Transfer Learning with Sub-word Tokenization on Downstream Applications
🖥 Github: hhttps://github.com/ginnm/proteinpretraining
📕 Paper: https://arxiv.org/pdf/2310.17415v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/peta-protein
⭐ Tasks: https://paperswithcode.com/task/language-modelling
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/peta-protein
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7❤1
Bilingual Corpus Mining and Multistage Fine-Tuning for Improving Machine Translation of Lecture Transcripts
🖥 Github: https://github.com/shyyhs/CourseraParallelCorpusMining
📕 Paper: https://arxiv.org/abs/2311.03696v1
🔥 Datasets: https://paperswithcode.com/dataset/aspec
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/aspec
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍6🔥3❤2
Quantized Distillation for Driver Activity Recognition
🖥 Github: https://github.com/calvintanama/qd-driver-activity-reco
📕 Paper: https://arxiv.org/pdf/2311.05970v1.pdf
🔥 Datasets: https://paperswithcode.com/dataset/drive-act
✨ Tasks: https://paperswithcode.com/task/activity-recognition
@ArtificialIntelligencedl
🔥 Datasets: https://paperswithcode.com/dataset/drive-act
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤6👍3🔥1
Inherently Interpretable Time Series Classification via Multiple Instance Learning (MILLET)
🖥 Github: https://github.com/jaearly/miltimeseriesclassification
📕 Paper: https://arxiv.org/pdf/2311.10049v1.pdf
✨ Tasks: https://paperswithcode.com/task/decision-making
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
🔥7👍3❤2
minimax: Efficient Baselines for Autocurricula in JAX
🖥 Github: https://github.com/facebookresearch/minimax
📕 Paper: https://arxiv.org/pdf/2311.12716v1.pdf
✨ Tasks: https://paperswithcode.com/task/decision-making
@ArtificialIntelligencedl
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
❤7👍2🔥2
♟ ChessVision - A dataset for logically coherent multi-label classification.
Набор данных с изображениями шахматных досок в различных позициях.
🖥 Github: https://github.com/espressovi/chessvisionchallenge
📕 Paper: https://arxiv.org/pdf/2311.12610v1.pdf
✨ Tasks: https://paperswithcode.com/task/classification-1
@ArtificialIntelligencedl
Набор данных с изображениями шахматных досок в различных позициях.
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7❤4🔥3
Infant Action Recognition
🖥 Github: https://github.com/ostadabbas/video-based-infant-action-recognition
📕 Paper: https://arxiv.org/pdf/2311.12300v1.pdf
✨ Tasks: https://paperswithcode.com/task/action-recognition-in-videos
🔥Datasets: https://paperswithcode.com/dataset/ntu-rgb-d
@ArtificialIntelligencedl
🔥Datasets: https://paperswithcode.com/dataset/ntu-rgb-d
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍7❤3
CSMeD: Citation Screening Meta-Dataset for systematic review automation evaluation
🖥 Github: https://github.com/wojciechkusa/systematic-review-datasets
📕 Paper: https://arxiv.org/pdf/2311.12474v1.pdf
✨ Tasks: https://paperswithcode.com/task/question-answering
🔥Datasets: https://paperswithcode.com/dataset/blurb
@ArtificialIntelligencedl
🔥Datasets: https://paperswithcode.com/dataset/blurb
@ArtificialIntelligencedl
Please open Telegram to view this post
VIEW IN TELEGRAM
👍6❤2