#few-shot learning
19 papers · 1 filter
A comparative analysis of automated techniques for security bug report identification
Muhammad Laiq
The paper compares various automated methods, including traditional machine‑learning models and large language models, for identifying security‑related bug reports, finding that th…
Dual-Path LLM Reasoning for Multimodal Few-Shot Knowledge Graph Completion
Jinlan Liu, Zhiying Tu, Yongchao Xing +5
The paper introduces DuPLeR, a dual-path reasoning framework that combines large language model priors with multimodal signals to improve few-shot and zero-shot knowledge graph com…
Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
Tianyan Deng, Yanxiong Li, Rui Gao +1
The paper proposes a two‑phase transductive method that refines class prototypes while down‑weighting unknown‑class audio samples, enabling few‑shot classification of known sounds…
Evaluating Prompt Scope and Demonstration Similarity in Local LLM Machine Translation
Mihael Arcan
The paper studies how different prompt scopes (single vs. family-wide) and demonstration selection methods (random, lexical, embedding similarity) affect the translation quality of…
Bi-Level Collaborative Learning for Few-Shot Scribble-Supervised Medical Image Segmentation
Xiang-Xiang Su, Yufan Ye, Yihang Zheng +2
The paper introduces a bi-level collaborative learning framework that combines a learnable superpixel model with a segmentation network to improve few-shot, scribble‑supervised med…
VQ-Touch: A Data-Efficient Tactile Generation Framework Across Sensors and Scenarios
Kailin Lyu, Long Xiao, Jianing Zeng +3
The paper presents VQ-Touch, a framework that efficiently generates high‑fidelity tactile images across different sensors and scenarios using a VQ‑GAN based representation and a di…