English
Related papers

Related papers: From Fact Overwriting to Knowledge Evolution: Caus…

200 papers

Knowledge editing aims to efficiently update Large Language Models (LLMs) by modifying specific knowledge without retraining the entire model. Among knowledge editing approaches, in-context editing (ICE) offers a lightweight solution by…

Computation and Language · Computer Science 2025-06-03 Changyue Wang , Weihang Su , Qingyao Ai , Yujia Zhou , Yiqun Liu

Knowledge distillation (KD) is a technique that compresses large teacher models by training smaller student models to mimic them. The success of KD in auto-regressive language models mainly relies on Reverse KL for mode-seeking and…

Computation and Language · Computer Science 2024-09-23 Jun Rao , Xuebo Liu , Zepeng Lin , Liang Ding , Jing Li , Dacheng Tao , Min Zhang

On-Policy Distillation (OPD) has gained wide attraction as an LLM post-training paradigm due to its effectiveness in improving capabilities without introducing model distribution drift, and consequently, regression in general tasks.…

Artificial Intelligence · Computer Science 2026-05-25 Aristotelis Lazaridis , Dylan Bates , Aman Sharma , Brian King , Vincent Lu , Jack FitzGerald

The potential of using a large language model (LLM) as a knowledge base (KB) has sparked significant interest. To manage the knowledge acquired by LLMs, we need to ensure that the editing of learned facts respects internal logical…

Computation and Language · Computer Science 2023-12-05 Zichao Li , Ines Arous , Siva Reddy , Jackie C. K. Cheung

Data-Free Knowledge Distillation (DFKD) is a promising task to train high-performance small models to enhance actual deployment without relying on the original training data. Existing methods commonly avoid relying on private data by…

Computer Vision and Pattern Recognition · Computer Science 2024-03-29 Yuzheng Wang , Dingkang Yang , Zhaoyu Chen , Yang Liu , Siao Liu , Wenqiang Zhang , Lihua Zhang , Lizhe Qi

On-policy reinforcement learning has become the dominant paradigm for reasoning alignment in large language models, yet its sparse, outcome-level rewards make token-level credit assignment notoriously difficult. On-Policy Distillation (OPD)…

Machine Learning · Computer Science 2026-04-14 Binbin Zheng , Xing Ma , Yiheng Liang , Jingqing Ruan , Xiaoliang Fu , Kepeng Lin , Benchang Zhu , Ke Zeng , Xunliang Cai

Large Language Models (LLMs) represent substantial intellectual and economic investments, yet their effectiveness can inadvertently facilitate model imitation via knowledge distillation (KD). In practical scenarios, competitors can distill…

Machine Learning · Computer Science 2025-10-21 Pingzhi Li , Zhen Tan , Mohan Zhang , Huaizhi Qu , Huan Liu , Tianlong Chen

Knowledge editing (KE) provides a lightweight alternative to repeated fine-tuning of LLMs. However, most existing KE methods target dense feed-forward layers, while modern LLMs increasingly adopt Mixture-of-Experts (MoE) architectures for…

Machine Learning · Computer Science 2026-05-19 Roman Maksimov , Vladimir Aletov , Dmitry Bylinkin , Daniil Medyakov , Vladimir Solodkin , Aleksandr Beznosikov

Current approaches of knowledge editing struggle to effectively propagate updates to interconnected facts. In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for…

Computation and Language · Computer Science 2024-02-01 Wenyue Hua , Jiang Guo , Mingwen Dong , Henghui Zhu , Patrick Ng , Zhiguo Wang

Growing efforts to improve knowledge distillation (KD) in large language models (LLMs) replace dense teacher supervision with selective distillation, which uses a subset of token positions, vocabulary classes, or training samples for…

Computation and Language · Computer Science 2026-02-03 Almog Tavor , Itay Ebenspanger , Neil Cnaan , Mor Geva

Since the advent of reasoning-based large language models, many have found great success from distilling reasoning capabilities into student models. Such techniques have significantly bridged the gap between reasoning and standard LLMs on…

Knowledge editing aims to modify outdated knowledge in language models efficiently while retaining their original capabilities. Mainstream datasets for knowledge editing are predominantly static and fail to keep in pace with the evolving…

Computation and Language · Computer Science 2026-04-24 Chenming Tang , Yutong Yang , Kexue Wang , Yunfang Wu

Self-distillation has emerged as an effective post-training paradigm for LLMs, often improving performance while shortening reasoning traces. However, in mathematical reasoning, we find that it can reduce response length while degrading…

Computation and Language · Computer Science 2026-05-21 Jeonghye Kim , Xufang Luo , Minbeom Kim , Sangmook Lee , Dohyung Kim , Jiwon Jeon , Dongsheng Li , Yuqing Yang

On-policy self-distillation, where a student is pulled toward a copy of itself conditioned on privileged context (e.g., a verified solution or feedback), offers a promising direction for advancing reasoning capability without a stronger…

Machine Learning · Computer Science 2026-05-13 Guobin Shen , Xiang Cheng , Chenxiao Zhao , Lei Huang , Jindong Li , Dongcheng Zhao , Xing Yu

On-policy self-distillation (self-OPD) densifies reinforcement learning with verifiable rewards (RLVR) by letting a policy teach itself under privileged context. We find that when this guidance spans the full response, all-token KL spends…

Artificial Intelligence · Computer Science 2026-05-12 Jiaxuan Wang , Xuan Ouyang , Zhiyu Chen , Yulan Hu , Zheng Pan , Xin Li , Lan-Zhe Guo

We propose a causal framework to explain the catastrophic forgetting in Class-Incremental Learning (CIL) and then derive a novel distillation method that is orthogonal to the existing anti-forgetting techniques, such as data replay and…

Artificial Intelligence · Computer Science 2021-03-09 Xinting Hu , Kaihua Tang , Chunyan Miao , Xian-Sheng Hua , Hanwang Zhang

Knowledge editing (KE) methods offer an efficient way to modify knowledge in large language models. Current KE evaluations typically assess editing success by considering only the edited knowledge without any preceding contexts. In…

Computation and Language · Computer Science 2025-06-03 Haewon Park , Gyubin Choi , Minjun Kim , Yohan Jo

Text Style Transfer (TST) seeks to alter the style of text while retaining its core content. Given the constraints of limited parallel datasets for TST, we propose CoTeX, a framework that leverages large language models (LLMs) alongside…

Computation and Language · Computer Science 2024-05-07 Chiyu Zhang , Honglong Cai , Yuezhang , Li , Yuexin Wu , Le Hou , Muhammad Abdul-Mageed

Knowledge Distillation is a technique which aims to utilize dark knowledge to compress and transfer information from a vast, well-trained neural network (teacher model) to a smaller, less capable neural network (student model) with improved…

Computer Vision and Pattern Recognition · Computer Science 2022-01-28 Fahad Rahman Amik , Ahnaf Ismat Tasin , Silvia Ahmed , M. M. Lutfe Elahi , Nabeel Mohammed

In-context knowledge editing (IKE) enables efficient modification of large language model (LLM) outputs without parameter changes and at zero-cost. However, it can be misused to manipulate responses opaquely, e.g., insert misinformation or…

Computation and Language · Computer Science 2025-04-11 Paul Youssef , Zhixue Zhao , Jörg Schlötterer , Christin Seifert