one shot learning
30 articles about one shot learning in AI news
Beyond Flat Space: How Hyperbolic Geometry Solves AI's Few-Shot Learning Bottleneck
Researchers propose Hyperbolic Flow Matching (HFM), a novel approach using hyperbolic geometry to dramatically improve few-shot learning. By leveraging the exponential expansion of Lorentz manifolds, HFM prevents feature entanglement that plagues traditional Euclidean methods, achieving state-of-the-art results across 11 benchmarks.
One Policy to Rule Them All: AI Robot Masters Unseen Tools with Zero-Shot Generalization
Researchers have developed a single robot policy capable of manipulating diverse, never-before-seen tools using sim-to-real reinforcement learning. The system achieves zero-shot generalization across 24 tasks, 12 objects, and 6 tool categories without object-specific training.
DART: One-Shot Robot Adaptation via Weight Space Arithmetic
DART from Seoul National University adapts robot policies with one demonstration using weight space arithmetic, achieving 73% success on unseen domain shifts.
Tencent's UI-Mate: One-Shot GUI Agent Re-Plans Live
Tencent released UI-Mate, an open-weight GUI agent that learns from one demo and re-plans live, but lacks technical specs.
Tencent's UI-Mate-27B Hits 77.0 OSWorld, Learns From One Demo
Tencent's UI-Mate-27B scores 77.0 OSWorld-Verified, 66.2 WindowsAgentArena, learning GUI tasks from one demo. Released on Hugging Face.
LLM Schema-Adaptive Method Enables Zero-Shot EHR Transfer
Researchers propose Schema-Adaptive Tabular Representation Learning, an LLM-driven method that transforms structured variables into semantic statements. It enables zero-shot alignment across unseen EHR schemas and outperforms clinical baselines, including neurologists, on dementia diagnosis tasks.
AI Learns to Use Tools Without Expensive Training: The Rise of In-Context Reinforcement Learning
Researchers have developed In-Context Reinforcement Learning (ICRL), a method that teaches large language models to use external tools through demonstration examples during reinforcement learning. This approach eliminates costly supervised fine-tuning while enabling models to gradually transition from few-shot to zero-shot tool usage capabilities.
Alibaba's Qwen-RobotNav Unifies Robot Navigation in One 2B-8B Model
Alibaba's Qwen-RobotNav unifies VLN, ObjectNav, tracking, and autonomous driving in a 2B-8B model, deploying zero-shot to quadruped robots via a configurable observation protocol.
Perplexity AI Launches Live Personal Money Analyzer via Plaid
Perplexity AI has integrated with Plaid to transform its finance Q&A feature into a live personal money analyzer, allowing users to query their own transaction data. This move directly challenges incumbents in the AI-powered personal finance space.
Bones Studio Demos Motion-Capture-to-Robot Pipeline for Home Tasks
Bones Studio released a demo showing its 'Captured → Labeled → Transferred' pipeline. It uses optical motion capture to record human tasks, then transfers the data for a humanoid robot to replicate the actions in simulation.
Stanford's EgoNav Trains Robot Navigation on 5 Hours of Human Video, Enables Zero-Shot Control of Unitree G1
Stanford's EgoNav system uses a 5-hour egocentric video walk of campus to train a diffusion model that enables zero-shot navigation for a Unitree G1 humanoid robot, eliminating the need for robot-specific training data.
Momentum-Consistency Fine-Tuning (MCFT) Achieves 3.30% Gain in 5-Shot 3D Vision Tasks Without Adapters
Researchers propose MCFT, an adapter-free fine-tuning method for 3D point cloud models that selectively updates encoder parameters with momentum constraints. It outperforms prior methods by 3.30% in 5-shot settings and maintains original inference latency.
AI2's MolmoWeb: Open 8B-Parameter Web Agent Navigates Using Screenshots, Challenges Proprietary Systems
The Allen Institute for AI released MolmoWeb, a fully open web agent that operates websites using only screenshots. The 8B-parameter model outperforms other open models and approaches proprietary performance, with all training data and weights publicly released.
Meta's V-JEPA 2.1 Achieves +20% Robotic Grasp Success with Dense Feature Learning from 1M+ Hours of Video
Meta researchers released V-JEPA 2.1, a video self-supervised learning model that learns dense spatial-temporal features from over 1 million hours of video. The approach improves robotic grasp success by ~20% over previous methods by forcing the model to understand precise object positions and movements.
FedShare: A New Framework for Federated Recommendation with Personalized Data Sharing and Unlearning
Researchers propose FedShare, a federated learning framework for recommender systems that allows users to dynamically share data for better performance and request its removal via efficient 'unlearning', addressing a key privacy-performance trade-off.
NVIDIA's Kimi-K2.5 Eagle Head: Supercharging Moonshot's Reasoning with Speculative Decoding
NVIDIA has released the Kimi-K2.5 Eagle head on Hugging Face, implementing Eagle-3 speculative decoding to dramatically accelerate inference for Moonshot's reasoning models. This breakthrough promises blazing-fast performance while maintaining accuracy.
Beyond Catastrophic Forgetting: AI Research Pioneers Self-Regulating Neural Architectures
Two breakthrough papers introduce Non-Interfering Weight Fields for zero-forgetting learning and objective-free learning systems that self-regulate based on internal dynamics. These approaches could fundamentally change how AI models acquire and retain knowledge.
Ring-Zero Trains 1T-Parameter Model via Reinforcement Learning
Ring-Zero scales RL with verifiable rewards to 1T parameters, revealing emergent reasoning like self-verification and context anxiety.
Ethan Mollick: Current AI Tooling Is a 'Substitute' for Continual Learning
Ethan Mollick observes that the entire ecosystem of prompts, skill files, and retrieval tools is a patch for AI's inability to learn continually. If solved, this would rapidly obsolete much current tooling.
Google's TimesFM: 200M-Param Foundation Model for Zero-Shot Time Series
Google released TimesFM, a 200M-parameter foundation model for time series forecasting that works without training on user data. It's now available open-source and as a product inside BigQuery.
Google Open-Sources TimesFM: A 100B-Point Time Series Foundation Model for Zero-Shot Forecasting
Google has open-sourced TimesFM, a foundation model for time series forecasting trained on 100 billion real-world time points. It requires no dataset-specific training and can generate predictions instantly for domains like traffic, weather, and demand.
Microsoft Open-Sources VALL-E 2: A Zero-Shot TTS Model Achieving Human Parity in Speech Naturalness
Microsoft Research has open-sourced VALL-E 2, a neural codec language model for text-to-speech that achieves human parity in naturalness. It uses a novel 'Repetition-Aware Sampling' method to eliminate word repetition, a common failure mode in prior models.
Google DeepMind's 'Learning Through Conversation' Paper Shows LLMs Can Improve with Real-Time Feedback
Google DeepMind researchers have published a paper demonstrating that large language models can be trained to learn and improve their responses during a conversation by incorporating user feedback, moving beyond static pre-training.
EvoSkill: How AI Agents Are Learning to Teach Themselves New Skills
Researchers have developed EvoSkill, a self-evolving framework where AI agents automatically discover and refine their own capabilities through failure analysis. The system improves performance by up to 12% on complex tasks and demonstrates skill transfer between different domains.
VoteGCL: A Novel LLM-Augmented Framework to Combat Data Sparsity in
A new paper introduces VoteGCL, a framework that uses few-shot LLM prompting and majority voting to create high-confidence synthetic data for graph-based recommendation systems. It integrates this data via graph contrastive learning to improve accuracy and mitigate bias, outperforming existing baselines.
New Research Shows Pre-Aligned Multi-Modal Models Advance 3D Shape Retrieval from Images
A new arXiv paper demonstrates that pre-aligned image and 3D shape encoders, combined with hard contrastive learning, achieve state-of-the-art performance for image-based shape retrieval. This enables zero-shot retrieval without database-specific training.
CLIPoint3D Bridges the 3D Reality Gap: How Language Models Are Revolutionizing Point Cloud Adaptation
Researchers have developed CLIPoint3D, a novel framework that leverages frozen CLIP backbones for few-shot unsupervised 3D point cloud domain adaptation. The approach achieves 3-16% accuracy gains over conventional methods while dramatically improving efficiency by avoiding heavy trainable encoders.
Production Deployment Patterns for AI Agent Systems: From Prototype to Scale
The article presents CI/CD, monitoring, rollback, and scaling patterns for AI agent production deployments from a SaaS practitioner. It emphasizes treating multi-agent workflows as atomic units, using OpenTelemetry tracing, and implementing circuit breakers for resilience.
Claude Now Tutors Kids for Free, Matching $100/hr Private Lessons
Claude can now teach kids any school subject like a $100/hour private tutor from Khan Academy, for free. This brings high-quality, personalized AI tutoring to anyone with internet access.
Kimi 2.6 Thinking Shows Promise as Open Weights Model, Lags Behind Closed SoTA
An initial evaluation of Moonshot AI's Kimi 2.6 Thinking model finds it generates extensive reasoning traces but delivers only 'okay-ish' results on creative and coding tasks, highlighting the persistent open vs. closed model gap.