AI Briefing#zero-shot
The latest AI and developer news about #zero-shot, with the original source and a short summary.
The latest AI and developer news about #zero-shot, with the original source and a short summary.
Benchmarking across 29 datasets shows GLM-5.3-Flash achieves statistical parity with the specialized Jev model for intent classification without fine-tuning, though at higher latency and cost.
Google Research has released TimesFM-3, a 330-million-parameter time-series foundation model that predicts multivariate data and covariates in a single forward pass.
Google Research has unveiled TabFM, a model that processes tabular data without separate fine-tuning.
Google has released TabFM, a model that can make predictions for classification and regression tasks on tabular data without separate training.
Scenema Audio has released weights for a zero-shot, emotion-decoupled voice generation model.
OpenAI privacy-filter led in CPU speed, but its strict F1 was lower due to tokenizer offset issues.
BabyZWM, trained solely on a single child's visual experience, matched SOTA on multiple visual-cognitive tasks.
LAVE is proposed as a metric that uses an LLM to evaluate semantic similarity instead of the existing string-matching approach.
Pollen-Vision, an open-source library integrating Zero-Shot vision models to help robots perceive and manipulate objects, has been released.
This introduces how CLIPSeg, a CLIP-based zero-shot image segmentation model, works and how to use it.