#NLProc #UKPLab #NLP #LLMs #AI
#MultiDocumentReasoning #ZeroShotLearning #Evaluation #OpenScience @cs-tudarmstadt.bsky.social
Latest posts tagged with #ZeroShotLearning on Bluesky
#NLProc #UKPLab #NLP #LLMs #AI
#MultiDocumentReasoning #ZeroShotLearning #Evaluation #OpenScience @cs-tudarmstadt.bsky.social
Ever wonder how new foundation models are rewriting time‑series forecasting? From Chronos‑2’s zero‑shot tricks to Hugging Face’s pre‑trained gems, the future of ML forecasting just got a serious upgrade. Dive in! #TimeSeriesFoundation #ZeroShotLearning #Chronos2
🔗 aidailypost.com/news/five-em...
We tested GPT-4, Llama-2, and more on symbolic puzzles—see why even the strongest LLMs fail without fine-tuning. #zeroshotlearning
Can LLMs solve math? This study explores puzzles, synthetic data, and fine-tuning to push AI’s limits in reasoning and extrapolation. #zeroshotlearning
Fine-tuning boosts AI reasoning: models trained on 100M samples achieve higher pass@1 rates in puzzle-solving across in-distribution and OOD tests. #zeroshotlearning
A new dataset turns arithmetic puzzles into a benchmark for AI, testing LLaMA’s reasoning with LoRA fine-tuning and OOD evaluation. #zeroshotlearning
Fine-tuning LLMs with synthetic data boosts multi-step mathematical reasoning and improves zero-shot performance on novel benchmarks. #zeroshotlearning
zero-shot prompt: “map all whataburgers in texas” and nailed it, catch the full video exclusively on collide.io #zeroshotlearning #llmdemo #codingwithai
The TechBeat: What 300GB of AI Research Reveals About the True Limits of “Zero-Shot” Intelligence (7/12/2025) #Technology #EmergingTechnologies #ArtificialIntelligence #AIResearch #ZeroShotLearning #EmergingTech
New research shows how T2I models like Stable Diffusion struggle with rare prompts, revealing key accuracy gaps across head and tail concepts. #zeroshotlearning
Why RAM++ beats CLIP for fine-grained image concept detection, with a focus on threshold tuning and identifying misaligned image-text pairs. #zeroshotlearning
Concept frequency improves zero-shot AI performance across image and text domains, confirmed through detailed dataset analysis and cleaning methods. #zeroshotlearning
Concept frequency in training data predicts zero-shot accuracy in T2I models like Stable Diffusion, especially when generating images of public figures. #zeroshotlearning
Concept frequency strongly predicts zero-shot AI performance across multiple prompting styles and six retrieval metrics, study confirms. #zeroshotlearning
Multimodal models struggle with long-tail concepts. This study analyzes 34 models and 300GB of data to reveal key limitations in zero-shot generalization. #zeroshotlearning
Long-tailed data in large-scale AI datasets affects model performance. This article analyzes the root causes and implications for future model training. #zeroshotlearning
New study reveals why AI models underperform on rare concepts using the “Let It Wag!” dataset of long-tail categories in classification and generation tasks.
#zeroshotlearning
New findings reveal long-tailed distributions, image-text misalignment, and consistent concept patterns across major AI pretraining datasets. #zeroshotlearning
Concept frequency reliably predicts AI performance—even when similar samples are removed or synthetic data is used for pretraining. #zeroshotlearning
Pretraining frequency strongly predicts zero-shot performance in multimodal models across classification, retrieval, and generative tasks. #zeroshotlearning
Learn how researchers quantify and align concepts across images and text in AI pretraining datasets using tagging, NLP, and model-based analysis. #zeroshotlearning
A study finds multimodal AI models like CLIP need exponential data for linear gains—raising questions about zero-shot generalization claims. #zeroshotlearning
Precise Zero-Shot Pointwise Ranking with LLMs through Post-Aggregated
Global Context Information
Chen Xu, Jintao Tang et al.
Paper
Details
#ZeroShotLearning #LargeLanguageModels #PointwiseRanking
Unlike TS FMs, DynaMix exhibits #ZeroShotLearning of long-term stats of unseen DS, incl. attractor geometry & power spectrum, w/o *any* re-training, just from a context signal.
It does so with only 0.1% of the parameters of Chronos & 10x faster inference times than the closest competitor.
(2/6)
A full reading of the recent paper from the researchers at Tsinghua University ,Penn State University, and #BeijingInstituteforGeneralArtificialIntelligence
#AIpaper #AbsoluteZero #ReinforcementLearning #AIResearch #LLMs #ZeroShotLearning #deeplearning
youtu.be/gmUz2bROrLk
ICPSR Summer Program Workshop Announcement: Research with Big Data: Data Harvesting and Artificial Intelligence, July 21-25, Online!
Learn about the data access landscape, develop code to collect data, get an overview of AI in computational social science research, and learn about zero-shot and few-shot learning with large language models. For more details: myumi.ch/Z2k2d
#SumProg25 #ICPSR #DataAccess #ZeroShotLearning
TxGNN improves drug repurposing by predicting treatments for rare diseases with no approved therapies 🔬💊📊 www.news-medical.net/news/2024092... #DrugRepurposing #AI #ZeroShotLearning #TxGNN #MedicalInnovation #RareDiseases #PrecisionMedicine
Innovative Humanoid Robot Alter3 Uses GPT-4 to Execute Detailed Commands.
See here - techchilli.com/robotics/inn...
#GPT4 #Alter3 #HumanoidRobot #AI #Robotics #Innovation #ZeroShotLearning #TechAdvancement #FutureOfAI #AIinteraction #RobotCapabilities #TechRevolution
Künstliche Intelligenz lernt nur durch Sprachanweisung unbekannte Aufgaben
#KünstlicheIntelligenz #NeuronaleNetze #ZeroShotLearning #SprachverarbeitungImGehirn #KognitivePsychologie #artificalintelligence #AI #NLP
kinews24.de/kuenstliche-...