English
Related papers

Related papers: SIGMA: Scalable Spectral Insights for LLM Model Co…

200 papers

Text-driven image editing has advanced rapidly, but reliably localizing these manipulations requires image manipulation localization (IML) models trained on large pixel-annotated datasets, and there is still no low-cost way to obtain such…

Computer Vision and Pattern Recognition · Computer Science 2026-05-28 Peiyu Zhuang , Jianquan Yang , Haodong Li , Zhuoying Cai , Ruitao Xie , Jishen Zeng , Baoying Chen , Jiwu Huang , Xiaochun Cao

Foundation models, such as large language models (LLMs), are powerful but often require customization before deployment to satisfy practical constraints such as safety, privacy, and task-specific requirements, leading to "constrained"…

Recent works leveraging Graph Neural Networks to approach graph matching tasks have shown promising results. Recent progress in learning discrete distributions poses new opportunities for learning graph matching models. In this work, we…

Machine Learning · Computer Science 2021-09-14 Linfeng Liu , Michael C. Hughes , Soha Hassoun , Li-Ping Liu

Large Language Models (LLMs) like LLaMA, Mistral, and Gemma are increasingly used in decision-critical domains such as healthcare, law, and finance, yet their reliability remains uncertain. They often make overconfident errors, degrade…

Computation and Language · Computer Science 2026-01-01 Rohit Kumar Salla , Manoj Saravanan , Shrikar Reddy Kota

Understanding how feature representations evolve across layers in large language models (LLMs) is key to improving their interpretability and robustness. While recent studies have identified critical layers linked to specific functions or…

Machine Learning · Computer Science 2025-06-06 Xuyuan Liu , Lei Hsiung , Yaoqing Yang , Yujun Yan

The integration of speech into Large Language Models (LLMs) has substantially expanded their capabilities, but often at the cost of weakening their core textual competence. This degradation limits the ability of speech-enabled LLMs to fully…

Computation and Language · Computer Science 2025-09-30 Chao Wang , Rui-Chen Zheng , Yang Ai , Zhen-Hua Ling

Synthetic data has been increasingly used to train frontier generative models. However, recent studies raise key concerns that iteratively retraining a generative model on its self-generated synthetic data may keep deteriorating model…

Machine Learning · Statistics 2026-03-09 Bingji Yi , Qiyuan Liu , Yuwei Cheng , Haifeng Xu

Doubly-stochastic matrices (DSM) are increasingly utilized in structure-preserving deep architectures -- such as Optimal Transport layers and Sinkhorn-based attention -- to enforce numerical stability and probabilistic interpretability. In…

Machine Learning · Computer Science 2026-01-06 Yizhi Liu

Large Language Models (LLMs) are known for their performance, but we uncover a significant structural inefficiency: a phenomenon we term attention collapse. In many pre-trained decoder-style LLMs, the attention matrices in deeper layers…

Computation and Language · Computer Science 2026-02-17 Sunny Sanyal , Ravid Shwartz-Ziv , Alexandros G. Dimakis , Sujay Sanghavi

Scaling the amount of data used for supervied fine-tuning(SFT) does not guarantee the proportional gains in model performance, highlighting a critical need to understand what makes training samples effective. This work identifies two…

Artificial Intelligence · Computer Science 2025-10-29 Chengwei Wu , Li Du , Hanyu Zhao , Yiming Ju , Jiapu Wang , Tianyu Chen , Haoyi Zhou

As scaled language models (LMs) approach human-level reasoning capabilities, self-improvement emerges as a solution to synthesizing high-quality data corpus. While previous research has identified model collapse as a risk in…

Computation and Language · Computer Science 2025-10-28 Xiangchi Yuan , Chunhui Zhang , Zheyuan Liu , Dachuan Shi , Leyan Pan , Soroush Vosoughi , Wenke Lee

Large language models (LLMs) demonstrate strong performance across natural language processing tasks, yet undergo significant performance degradation when modified for deployment through quantization, pruning, or decoding strategy…

Computation and Language · Computer Science 2025-04-01 Ziyang Ma , Zuchao Li , Lefei Zhang , Gui-Song Xia , Bo Du , Liangpei Zhang , Dacheng Tao

Effective LLM training depends on predictable scaling of key quantities -- such as final loss and optimal hyperparameters -- with model and dataset size. Qiu et al. (2025) recently showed that this predictability can extend beyond scalars:…

Machine Learning · Computer Science 2026-03-03 Shane Bergsma , Bin Claire Zhang , Nolan Dey , Shaheer Muhammad , Gurpreet Gosal , Joel Hestness

State-space models (SSMs), particularly Mamba, emerge as an efficient Transformer alternative with linear complexity for long-sequence modeling. Recent empirical works demonstrate Mamba's in-context learning (ICL) capabilities competitive…

Machine Learning · Computer Science 2025-09-30 Jiarui Jiang , Wei Huang , Miao Zhang , Taiji Suzuki , Liqiang Nie

Training loss and throughput can hide distinct internal representation in language-model training. To examine these hidden mechanics, we use spectral measurements as practical and operational diagnostics. Using a controlled family of…

Machine Learning · Statistics 2026-05-08 Andy Zeyi Liu , Elliot Paquette , John Sous

Large Language Models (LLMs) have demonstrated remarkable capabilities, yet their scalability raises a critical question: Have we reached the scaling ceiling? This paper addresses this pivotal question by developing a unified theoretical…

Machine Learning · Computer Science 2024-12-24 Charles Luo

Training Large Language Models (LLMs) at ultra-low precision is critically impeded by instability rooted in the conflict between discrete quantization constraints and the intrinsic heavy-tailed spectral nature of linguistic data. By…

Machine Learning · Computer Science 2026-02-03 Junlin Huang , Wenyi Fang , Zhenheng Tang , Yuxin Wang , Xueze Kang , Yang Zheng , Bo Li , Xiaowen Chu

The widespread use of generative models has created a feedback loop, in which each generation of models is trained on data partially produced by its predecessors. This process has raised concerns about model collapse: A critical degradation…

Machine Learning · Statistics 2026-03-27 Daniel Barzilai , Ohad Shamir

Neural collapse is a highly symmetric geometric pattern of neural networks that emerges during the terminal phase of training, with profound implications on the generalization performance and robustness of the trained networks. To…

Machine Learning · Computer Science 2022-04-26 Wenlong Ji , Yiping Lu , Yiliang Zhang , Zhun Deng , Weijie J. Su

The evaluation of large language models (LLMs) relies heavily on standardized benchmarks. These benchmarks provide useful aggregated metrics for a given capability, but those aggregated metrics can obscure (i) particular sub-areas where the…

Computation and Language · Computer Science 2025-12-25 Matyas Bohacek , Nino Scherrer , Nicholas Dufour , Thomas Leung , Christoph Bregler , Stephanie C. Y. Chan