English
Related papers

Related papers: Disentangling MLP Neuron Weights in Vocabulary Spa…

200 papers

Large Language Models (LLMs) have demonstrated strong performance in handling complex tasks requiring both extensive knowledge and reasoning abilities. However, the existing LLM inference pipeline operates as an opaque process without…

Computation and Language · Computer Science 2025-05-16 Mingyu Jin , Weidi Luo , Sitao Cheng , Xinyi Wang , Wenyue Hua , Ruixiang Tang , William Yang Wang , Yongfeng Zhang

An ongoing challenge in neural information processing is: how do neurons adjust their connectivity to improve task performance over time (i.e., actualize learning)? It is widely believed that there is a consistent, synaptic-level learning…

Neural and Evolutionary Computing · Computer Science 2021-06-01 Aman Bhargava , Mohammad R. Rezaei , Milad Lankarany

The ability to control for the kinds of information encoded in neural representation has a variety of use cases, especially in light of the challenge of interpreting these models. We present Iterative Null-space Projection (INLP), a novel…

Computation and Language · Computer Science 2020-04-30 Shauli Ravfogel , Yanai Elazar , Hila Gonen , Michael Twiton , Yoav Goldberg

We show that MLP layers in transformer language models perform binary routing of continuous signals: the decision of whether a token needs nonlinear processing is well-captured by binary neuron activations, even though the signals being…

Machine Learning · Computer Science 2026-03-12 Peter Balogh

Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms…

Machine Learning · Computer Science 2022-01-13 Pengyu Cheng , Martin Renqiang Min , Dinghan Shen , Christopher Malon , Yizhe Zhang , Yitong Li , Lawrence Carin

Multimodal large language models (MLLMs) have shown impressive capabilities, yet they often struggle to effectively capture the fine-grained textual information within images crucial for accurate image translation. This often leads to a…

Computation and Language · Computer Science 2026-04-21 Bo Li , Ningyuan Deng , Tianyu Dong , Shaobo Wang , Shaolin Zhu , Lijie Wen

The releases of powerful open-weight large language models (LLMs) are often not accompanied by access to their full training data. Existing interpretability methods, particularly those based on activations, often require or assume…

Machine Learning · Computer Science 2026-04-22 Ziqian Zhong , Aditi Raghunathan

Large language models (LLMs) possess vast knowledge acquired from extensive training corpora, but they often cannot remove specific pieces of information when needed, which makes it hard to handle privacy, bias mitigation, and knowledge…

Machine Learning · Computer Science 2025-12-09 Yezi Liu , Hanning Chen , Wenjun Huang , Yang Ni , Mohsen Imani

The large memory requirements of deep neural networks limit their deployment and adoption on many devices. Model compression methods effectively reduce the memory requirements of these models, usually through applying transformations such…

Machine Learning · Computer Science 2017-11-15 Brandon Reagen , Udit Gupta , Robert Adolf , Michael M. Mitzenmacher , Alexander M. Rush , Gu-Yeon Wei , David Brooks

Rotary Position Embedding (RoPE) has shown strong performance in text-based Large Language Models (LLMs), but extending it to video remains a challenge due to the intricate spatiotemporal structure of video frames. Existing adaptations,…

Artificial Intelligence · Computer Science 2025-11-03 Zikang Liu , Longteng Guo , Yepeng Tang , Tongtian Yue , Junxian Cai , Kai Ma , Qingbin Liu , Xi Chen , Jing Liu

The increasing size and complexity of large language models (LLMs) have raised significant challenges in deployment efficiency, particularly under resource constraints. Post-training quantization (PTQ) has emerged as a practical solution by…

Computation and Language · Computer Science 2026-04-07 Han Liu , Haotian Gao , Changya Li , Feng Zhang , Xiaotong Zhang , Wei Wang , Hong Yu

Spatial reasoning focuses on locating target objects based on spatial relations in 3D scenes, which plays a crucial role in developing intelligent embodied agents. Due to the limited availability of 3D scene-language paired data, it is…

Computer Vision and Pattern Recognition · Computer Science 2026-03-27 Shengli Zhou , Minghang Zheng , Feng Zheng , Yang Liu

As Large Language Models (LLMs) become increasingly prevalent, their security vulnerabilities have already drawn attention. Machine unlearning is introduced to seek to mitigate these risks by removing the influence of undesirable data.…

Machine Learning · Computer Science 2026-01-06 Xiang Zhang , Kun Wei , Xu Yang , Jiahua Li , Su Yan , Cheng Deng

We explore a new approach for training neural networks where all loss functions are replaced by hard constraints. The same approach is very successful in phase retrieval, where signals are reconstructed from magnitude constraints and…

Machine Learning · Computer Science 2019-11-04 Veit Elser

Disentangling polysemantic neurons is at the core of many current approaches to interpretability of large language models. Here we attempt to study how disentanglement can be used to understand performance, particularly under weight…

Machine Learning · Computer Science 2025-02-27 Shashata Sawmya , Linghao Kong , Ilia Markov , Dan Alistarh , Nir Shavit

We propose an empirical approach centered on the spectral dynamics of weights -- the behavior of singular values and vectors during optimization -- to unify and clarify several phenomena in deep learning. We identify a consistent bias in…

Spatial functional organization is a hallmark of biological brains: neurons are arranged topographically according to their response properties, at multiple scales. In contrast, representations within most machine learning models lack…

Computation and Language · Computer Science 2025-10-22 Taha Binhuraib , Greta Tuckute , Nicholas Blauch

Despite the efficacy of Direct Preference Optimization (DPO) in aligning Large Language Models (LLMs), reward hacking remains a pivotal challenge. This issue emerges when LLMs excessively reduce the probability of rejected completions to…

Computation and Language · Computer Science 2025-08-26 Chenxu Yang , Ruipeng Jia , Mingyu Zheng , Naibin Gu , Zheng Lin , Siyuan Chen , Weichong Yin , Hua Wu , Weiping Wang

Deep learning based methods hold state-of-the-art results in image denoising, but remain difficult to interpret due to their construction from poorly understood building blocks such as batch-normalization, residual learning, and feature…

Image and Video Processing · Electrical Eng. & Systems 2021-03-09 Nikola Janjušević , Amirhossein Khalilian-Gourtani , Yao Wang

In this paper, we present a novel data-free method for merging neural networks in weight space. Differently from most existing works, our method optimizes for the permutations of network neurons globally across all layers. This allows us to…

Machine Learning · Computer Science 2024-10-31 Donato Crisostomi , Marco Fumero , Daniele Baieri , Florian Bernard , Emanuele Rodolà