paper-with-me

홈 › Papers

From Amateur to Master: Infusing Knowledge into LLMs via Automated Curriculum Learning

2025-10-30 · Nishit Neema, Srinjoy Mukherjee, Sapan Shah, Gokul Ramakrishnan, Ganesh Venkatesh arxiv

Large Language Models (LLMs) excel at general tasks but underperform in specialized domains like economics and psychology, which require deep, principled understanding. To address this, we introduce ACER (Automated Curriculum-Enhanced Regimen) that transforms generalist models into domain experts without sacrificing their broad capabilities. ACER first synthesizes a comprehensive, textbook-style curriculum by generating a table of contents for a subject and then creating question-answer (QA) pairs guided by Bloom's taxonomy. This ensures systematic topic coverage and progressively increasing difficulty. The resulting synthetic corpus is used for continual pretraining with an interleaved curriculum schedule, aligning learning across both content and cognitive dimensions. Experiments with Llama 3.2 (1B and 3B) show significant gains in specialized MMLU subsets. In challenging domains like microeconomics, where baselines struggle, ACER boosts accuracy by 5 percentage points. Across all target domains, we observe a consistent macro-average improvement of 3 percentage points. Notably, ACER not only prevents catastrophic forgetting but also facilitates positive cross-domain knowledge transfer, improving performance on non-target domains by 0.7 points. Beyond MMLU, ACER enhances performance on knowledge-intensive benchmarks like ARC and GPQA by over 2 absolute points, while maintaining stable performance on general reasoning tasks. Our results demonstrate that ACER offers a scalable and effective recipe for closing critical domain gaps in LLMs.

📄 PDF Abstract BibTeX arXiv:2510.26336

Code (0)

등록된 구현이 없습니다.

Tasks

Continual PretrainingTopic coverage

Similar Papers 제목 키워드 기반

ContextGPT: Infusing LLMs Knowledge into Neuro-Symbolic Activity Recognition Models

2024-03-11 · Luca Arrotta, Claudio Bettini, Gabriele Civitarese, Michele Fiori

Context-aware Human Activity Recognition (HAR) is a hot research area in mobile computing, and the most effective solutions in the literature are based on supervised deep learning models. However, the actual deployment o…

Activity RecognitionCommon Sense ReasoningHuman Activity RecognitionPrompt Engineering

Reliable and diverse evaluation of LLM medical knowledge mastery

2024-09-22 · Yuxuan Zhou, Xien Liu, Chen Ning, Xiao Zhang 외

Mastering medical knowledge is crucial for medical-specific LLMs. However, despite the existence of medical benchmarks like MedQA, a unified framework that fully leverages existing knowledge bases to evaluate LLMs' maste…

DiversityMedQA

Addressing LLM Diversity by Infusing Random Concepts

2026-01-26 · Pulin Agrawal, Prasoon Goyal arxiv

Large language models (LLMs) are known to produce outputs with limited diversity. In this work, we study whether infusing random concepts in the prompts can improve the diversity of the generated outputs. To benchmark th…

Mixture-of-Partitions: Infusing Large Biomedical Knowledge Graphs into BERT

2021-09-10 · EMNLP 2021 11 · Zaiqiao Meng, Fangyu Liu, Thomas Hikaru Clark, Ehsan Shareghi 외

Infusing factual knowledge into pre-trained models is fundamental for many knowledge-intensive tasks. In this paper, we proposed Mixture-of-Partitions (MoP), an infusion approach that can handle a very large knowledge gr…

Knowledge Graphs

KSAM: Infusing Multi-Source Knowledge into Dialogue Generation via Knowledge Source Aware Multi-Head Decoding

2022-05-01 · Findings (ACL) 2022 5 · Sixing Wu, Ying Li, Dawei Zhang, Zhonghai Wu

Knowledge-enhanced methods have bridged the gap between human beings and machines in generating dialogue responses. However, most previous works solely seek knowledge from a single source, and thus they often fail to obt…

DecoderDialogue GenerationDiversity