We investigate how Large Language Models (LLMs) distinguish between memorization and generalization at the neuron level. Through carefully designed tasks, we identify distinct neuron subsets responsible for each behavior. Experiments on both a GPT-2 model trained from scratch and a pretrained LLaMA-3.2 model fine-tuned with LoRA show consistent neuron-level specialization. We further demonstrate that inference-time interventions on these neurons can steer the model’s behavior toward memorization or generalization. To assess robustness, we evaluate intra-task and inter-task consistency, confirming that these neuron-behavior associations reflect generalizable patterns rather than dataset-specific artifacts. Our findings reveal modular structure in LLMs and enable controlling memorization and generalization behaviors at inference time.

This study introduces a concept-based methodology to optimize Retrieval-Augmented Generation (RAG) tasks by assessing dataset certainty using entropy-based metrics and concept extraction techniques. Unlike traditional methods focused on reducing LLM hallucinations or modifying data structures, this approach evaluates inherent knowledge uncertainty from an LLM perspective. By pre-processing documents with LLMs, the concept-based method significantly enhances precision in tasks demanding high accuracy, such as legal, finance, or formal document responses.

This paper presents a comprehensive approach for the Formosa Speech Recognition Challenge 2025 (FSR-2025), targeting automatic speech recognition (ASR) for the under-resourced Dapu and Zhao’an dialects of Taiwanese Hakka. Our method integrates data augmentation and robustness techniques, including SpecAugment, dialect-aware special tokens, text-to-speech (TTS) augmentation, noise/reverberation mixing, and speed perturbation, to mitigate data scarcity and domain mismatch. Experiments on the official FSR-2025 datasets show consistent improvements in both character error rate (CER) and word error rate (WER). Extensive ablation studies further confirm that each component contributes positively. These results offer a practical path toward robust ASR for under-resourced Hakka dialects and suggest broader applicability to other low-resource languages.