📝 Encoding Multi-Domain Scientific Papers by Ensembling Multiple CLS Tokens 📚🧠

"Proposes Multi2SPE -- it encourages each of multiple CLS tokens to learn diverse ways of aggregating token embeddings, then sums them up together to create a single vector representation." [gal30b+] 🤖 #CL #DL #LG

🔗 https://arxiv.org/abs/2309.04333v1 #arxiv

📝 Large-Scale Automatic Audiobook Creation 🔊👾🧠

"Leverages recent advances in neural text-to-speech and text summarization and allows users to customize an audiobook's speaking style and speed using a small amount of speech samples." [gal30b+] 🤖 #SD #AI #DC #DL #LG

🔗 https://arxiv.org/abs/2309.03926v1 #arxiv

📝 OldVisOnline: Curating a Dataset of Historical Visualizations 👋

"Proposes a workflow to scrape and process heterogeneous metadata for historical visualization images from seven online libraries and propose a semi-automatic labeling approach to distinguish visualizations from other artifacts." [gal30b+] 🤖 #HC #DL

⚙️ https://github.com/oldvis
🔗 https://arxiv.org/abs/2308.16053v1 #arxiv

📝 Saving Temporary Exhibitions in Virtual Environments: The Digital Renaissance of Ulisse Aldrovandi 🖼️

"To make the digital twin of the temporary exhibition "The Other Renaissance: Ulisse Aldrovandi and the Wonders of the World" accessible online." [gal30b+] 🤖 #GR #DL

🔗 https://arxiv.org/abs/2308.15920v1 #arxiv

Interesting new lecture by Christopher Potts & his team, Stanford: Natural Language Understanding (CS224U, Spring 2023):
- Domain adaptation for supervised sentiment
- Retrieval augmented in-context learning
- Advanced behavioral evaluation and analysis methods
#NLP methods
Resources:
Video lectures: https://www.youtube.com/playlist?list=PLoROMvodv4rOwvldxftJTmoR3kRcWkJBp
Course website: http://web.stanford.edu/class/cs224u/index.html
Code: https://github.com/cgpotts/cs224u/
#ai #llm #deeplearning #dl #transformers #lecture

Re Al momento dell'insediamento il #GovernoMeloni aveva un numero di infrazioni pendenti con l'#UE di 84, rispetto ad una media europea di 65.
Con il #DL approvato oggi si interviene sull'avvio a soluzione e chiusura di 13 procedure e si previene l'apertura di nuove procedure.

🐦🔗: https://n.respublicae.eu/RaffaeleFitto/status/1687048275226918912

📝 UnScientify: Detecting Scientific Uncertainty in Scholarly Full Text 📚👾

"UnScientify automatically labels and annotates scientific uncertainty in scholarly papers by using a combination of pattern matching, complex sentence checking, and authorial reference checking approaches." [gal30b+] 🤖 #CL #AI #DL

🔗 https://arxiv.org/abs/2307.14236v1 #arxiv

📝 Unsupervised Extraction of Local and Global Keywords From a Single Text 📚

"Based on the spatial distribution of words and the change of this distribution due to a random permutation of words, in the text of a literary work (e." [gal30b+] 🤖 #CL #DL

🔗 https://arxiv.org/abs/2307.14005v1 #arxiv

Best #GPU for #DL (Deep Learning)

https://timdettmers.com/2023/01/30/which-gpu-for-deep-learning/

📝 OpenGDA: Graph Domain Adaptation Benchmark for Cross-Network Learning 👾

"To provide a benchmark for evaluating Graph Domain Adaptation models, OpenGDA provides unified preprocessed datasets for different types of tasks (i) node classification, (ii) link prediction or (iii) graph classification." [gal30b+] 🤖 #AI #DL #SI

⚙️ https://github.com/Skyorca/OpenGDA
🔗 https://arxiv.org/abs/2307.11341v1 #arxiv

📝 Classification of Visualization Types and Perspectives in Patents 🔭👾💿🧠

"Adopts state-of-the-art deep learning methods including transformers for the classification of visualization types and perspectives in patent images with a new patent dataset that contains a larger variety of visualization types and perspectives." [gal30b+] 🤖 #CV #AI #DL #IR #LG

⚙️ https://github.com/TIBHannover/PatentImageClassification
🔗 https://arxiv.org/abs/2307.10471v1 #arxiv

📝 Testing for Reviewer Anchoring in Peer Review: A Randomized Controlled Trial 👋

"Investigates whether reviewers anchor to their original scores when updating their reviews, which serves as a potential explanation for the lack of updates in reviewer scores (Figure 1)." [gal30b+] 🤖 #HC #DL

⚙️ https://github.com/theryanl/
🔗 https://arxiv.org/abs/2307.05443v1 #arxiv

📝 Can We Mathematically Spot Possible Manipulation of Results in Research Manuscripts Using Benford's Law? 💿

"An automatic tool that detects manipulated results using a statistical phenomenon (Benford's law) on commonly employed analyses in academic manuscripts, thus reducing the need for the raw data itself." [gal30b+] 🤖 #IR #DL

🔗 https://arxiv.org/abs/2307.01742v1 #arxiv

📝 Predictive Patentomics: Forecasting Innovation Success and Valuation with ChatGPT 🧠👾📚

"ChatGPT uses state-of-the-art deep learning techniques to process the unstructured text of 34,000,000 patents and their associated metadata and produce embeddings that are rich representations of the quality, potential impact, and other nuances of inventions." [gal30b+] 🤖 #LG #AI #CL #DL

🔗 https://arxiv.org/abs/2307.01202v1 #arxiv

📝 Multimodal Search on Iconclass Using Vision-Language Pre-Trained Models 💿👾

"The search engine is developed using a pre-trained CLIP vision-language model that is used to encode concepts in Iconclass as well as images and short textual descriptions associated with them." [gal30b+] 🤖 #IR #AI #DL

⚙️ https://github.com/ISE-FIZKarlsruhe/iconclass
🔗 https://arxiv.org/abs/2306.16529v1 #arxiv

📝 Weakly Supervised Multi-Label Classification of Full-Text Scientific Papers 📚

"A network-aware contrastive fine-tuning module and a hierarchy-aware aggregation module are designed to leverage the cross-paper network structure and the in-paper hierarchy structure of each paper, respectively." [gal30b+] 🤖 #CL #DL

🔗 https://arxiv.org/abs/2306.14003v1 #arxiv

📝 Has the Machine Learning Review Process Become More Arbitrary as the Field Has Grown? The NeurIPS 2021 Consistency Experiment 🧠

"We randomly selected a subset of NeurIPS (2021) submissions to be assigned to a second committee for review, and then measured the consistency between the two committees." [gal30b+] 🤖 #LG #DL

🔗 https://arxiv.org/abs/2306.03262v1 #arxiv

📝 DeepScribe: Localization and Classification of Elamite Cuneiform Signs via Deep Learning 🔭💿

"Develops DeepScribe, a modular computer vision pipeline that can analyze photos of cuneiform tablets from the Achaemenid period and provide useful transliteration suggestions to researchers." [gal30b+] 🤖 #CV #DL #IR

⚙️ http://github.com/edwardclem/deepscribe
🔗 https://arxiv.org/abs/2306.01268v1 #arxiv

Ever wondered what the brain of a tumor patient looked like before they developed the disease? This question keeps doctors awake at night. Help them by joining the BraTS Inpainting Challenge!
https://synapse.org/#!Synapse:syn51156910/wiki/622357

#AI #MachineLearning #DeepLearning #computervision #glioma #ml #dl

📝 Forgotten Knowledge: Examining the Citational Amnesia in NLP 📚

"Studies NLP papers published between 1990 and June 2019 and find that 82% of papers cite at least one paper that was published within 5 years before the target paper." [gal30b+] 🤖 #CL #DL

⚙️ https://github.com/iamjanvijay/
🔗 https://arxiv.org/abs/2305.18554v1 #arxiv