Lune

EMNLP Findings2024

Unleashing the Potential of Large Language Models through Spectral Modulation

Peng Sun, Yao Zhu, Yunjian Zhang, Xiu Yan, Zizhe Wang, Xiangyang Ji

Abstract

Large Language Models (LLMs) have demonstrated impressive capabilities across various domains, garnering significant attention from both academia and industry. However, enhancing the performance of LLMs typically requires scaling up model sizes or fine-tuning with additional datasets, which results in substantial computational costs. This paper poses an intriguing question: Can we improve the performance of LLMs without additional training? Drawing inspiration from signal processing principles, which suggest that noise often resides in high-frequency components while low-frequency components carry the essence of signals, we propose uncovering untapped potential in LLMs from a frequency perspective. We hypothesize that the high-frequency components in the weight matrices of LLMs' linear layers may conceal noise that interferes with predictive accuracy. Therefore, we propose conducting spectral modulation in the parameter space of LLMs, which can seamlessly integrate with various models in a plug-and-play manner. Extensive experiments have demonstrated the superiority of our approach, with spectral modulation yielding an average performance improvement of up to 10.12%.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

5 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • LoRA: Low-Rank Adaptation of Large Language Models cites Language Models are Few-Shot Learners
  • Learning Transferable Visual Models From Natural Language Supervision cites Language Models are Few-Shot Learners
  • Sliding-Window Merging for Compacting Patch-Redundant Layers in LLMs cites LoRA: Low-Rank Adaptation of Large Language Models
  • FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts cites LoRA: Low-Rank Adaptation of Large Language Models
  • FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts cites Language Models are Few-Shot Learners
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers5

  1. FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-ExpertsHeming Zou, Yunliang Zang, Wutong Xu, Yao Zhu et al.NeurIPS 2025 · 38 citations
  2. HypoChainer: A Collaborative System Combining LLMs and Knowledge Graphs for Hypothesis-Driven Scientific DiscoveryHaoran Jiang, Shaohan Shi, Yunjie Yao, Chang Jiang et al.IEEE VIS 2025 · 4 citations
  3. Sliding-Window Merging for Compacting Patch-Redundant Layers in LLMsXuan Ding, Rui Sun, Yunjian Zhang, Xiu Yan et al.AAAI 2026 · 4 citations
  4. Pruning as a Cooperative Game: Surrogate-Assisted Layer Contribution Estimation for Large Language ModelsXuan Ding, Pengyu Tong, Ranjie Duan, Yunjian Zhang et al.ICLR 2026 · 4 citations
  5. MGT-Prism: Enhancing Domain Generalization for Machine-Generated Text Detection via Spectral AlignmentShengchao Liu, Xiaoming Liu, Chengzhengxu Li, Zhaohan Zhang et al.AAAI 2026 · 1 citation

Builds on13

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
  3. LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
  4. TruthfulQA: Measuring How Models Mimic Human FalsehoodsStephanie Lin, Jacob Hilton, Owain EvansACL 2022 · 3,228 citations
  5. FcaNet: Frequency Channel Attention NetworksZequn Qin, Pengyi Zhang, Fei Wu, Xi LiICCV 2021 · 1,049 citations

Related papers

  1. The Truth is in There: Improving Reasoning in Language Models with Layer-Selective Rank ReductionPratyusha Sharma, Jordan T. Ash, Dipendra MisraICLR 2024 · 135 citations
  2. Language-guided Frequency Modulation for Large Vision-Language ModelsShuyi Ouyang, Gongfan Fang, Xinyin Ma, Yen-Wei Chen et al.CVPR 2026
  3. Interpreting and Improving Large Language Models in Arithmetic CalculationWei Zhang, Chaoqun Wan, Yonggang Zhang, Yiu-ming Cheung et al.ICML 2024 · 47 citations
  4. When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language ModelsHaoran You, Yichao Fu, Zheng Wang, Amir Yazdanbakhsh et al.ICML 2024 · 9 citations
  5. SPECTRA: Faster Large Language Model Inference with Optimized Internal and External SpeculationNguyen-Khang Le, Truong Dinh Do, Le-Minh NguyenACL 2025
Dusk over the sea between two cliffs drawn in fine vertical lines