SOTAVerified

Novel Concepts

Measures the ability of models to uncover an underlying concept that unites several ostensibly disparate entities, which hopefully would not co-occur frequently. This provides a limited test of a model's ability to creatively construct the necessary abstraction to make sense of a situation that it cannot have memorized in training.

Source: BIG-bench

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 26–50 of 158 papers

TitleStatusHype
Online Task-Free Continual Generative and Discriminative Learning via Dynamic Cluster MemoryCode1
Grounding Descriptions in Images informs Zero-Shot Visual RecognitionCode1
CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual LearningCode1
IFSeg: Image-free Semantic Segmentation via Vision-Language ModelCode1
PersonaMagic: Stage-Regulated High-Fidelity Face Customization with Tandem EquilibriumCode1
Dynamic Few-Shot Visual Learning without ForgettingCode1
Link-Context Learning for Multimodal LLMsCode1
XtarNet: Learning to Extract Task-Adaptive Representation for Incremental Few-Shot LearningCode1
Few-Shot Class-Incremental Learning via Class-Aware Bilateral DistillationCode1
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object InteractionsCode1
Decomposed Soft Prompt Guided Fusion Enhancing for Compositional Zero-Shot LearningCode1
Language-Informed Visual Concept LearningCode1
Beneath Surface Similarity: Large Language Models Make Reasonable Scientific Analogies after Structure AbductionCode0
Variational Prototype Replays for Continual LearningCode0
A Robust, Efficient Predictive Safety FilterCode0
Neural Concept Formation in Knowledge GraphsCode0
Can Vision Language Models Learn from Visual Demonstrations of Ambiguous Spatial Reasoning?Code0
Multi-level Semantic Feature Augmentation for One-shot LearningCode0
NeSyCoCo: A Neuro-Symbolic Concept Composer for Compositional GeneralizationCode0
Knowledge Graph Transfer Network for Few-Shot RecognitionCode0
L3DMC: Lifelong Learning using Distillation via Mixed-Curvature SpaceCode0
A Commonsense Reasoning Framework for Explanatory Emotion Attribution, Generation and Re-classificationCode0
BOWLL: A Deceptively Simple Open World Lifelong LearnerCode0
Learning like a Child: Fast Novel Visual Concept Learning from Sentence Descriptions of ImagesCode0
Deep Compositional Captioning: Describing Novel Object Categories without Paired Training DataCode0
Show:102550
← PrevPage 2 of 7Next →

No leaderboard results yet.