语言大模型 / LLM

COSTAR-A: A prompting framework for enhancing Large Language Model performance on Point-of-View questions

Nzubechukwu C. Ohalete, Kevin B. Gittner, Lauren M. Matheny

专题命中 效率与部署 ：large language model(title,abstract)；language model(title,abstract)；prompting(title,abstract)；分类 cs.CL

Comments 20 pages, 2 figures

Countermind: A Multi-Layered Security Architecture for Large Language Models

Dominik Schwarz

机构 * Independent Researcher（独立研究者） 纠错

专题命中 效率与部署 ：large language model(title,abstract)；language model(title,abstract)；LLM(abstract,comments)；分类 cs.AI

Comments 33 pages, 3 figures, 6 tables. Keywords: LLM security; defense-in-depth; prompt injection; activation steering; multimodal sandbox; threat modeling

Asking Clarifying Questions for Preference Elicitation With Large Language Models

Ali Montazeralghaem, Guy Tennenholtz, Craig Boutilier, Ofer Meshi

机构 * Google（谷歌） 纠错

专题命中 效率与部署 ：large language model(title,abstract)；language model(title,abstract)；LLM(abstract)；分类 cs.AI

Laminar: A Scalable Asynchronous RL Post-Training Framework

Guangming Sheng, Yuxuan Tong, Borui Wan, Wang Zhang, Chaobo Jia, Xibin Wu, Yuqi Wu, Xiang Li, Chi Zhang, Yanghua Peng, Haibin Lin, Xin Liu, Chuan Wu

机构 * The University of Hong Kong（香港大学） ； ByteDance Seed（字节跳动种子） 纠错

专题命中 效率与部署 ：post-training(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Task-Aware Reduction for Scalable LLM-Database Systems

Marcus Emmanuel Barnes, Taher A. Ghaleb, Safwat Hassan

机构 * Department of Computer Science Trent University Peterborough, Canada（计算机科学系特伦特大学彼得伯勒，加拿大） ； Faculty of Information University of Toronto Toronto, Canada（信息学院多伦多大学多伦多，加拿大） 纠错

专题命中 效率与部署 ：LLM(title,abstract)；language model(abstract,comments)；large language model(abstract)；分类 cs.CL

Comments Preprint. Accepted for presentation at the Workshop on Language Models and Databases (LMD), co-located with CASCON 2025 (IEEE). The final version will appear in IEEE Xplore

CARVQ: Corrective Adaptor with Group Residual Vector Quantization for LLM Embedding Compression

Dayin Gou, Sanghyun Byun, Nilesh Malpeddi, Gabrielle De Micheli, Prathamesh Vaste, Jacob Song, Woo Seong Chung

机构 * LG Electronics USA（LG电子美国公司） 纠错

专题命中 效率与部署 ：LLM(title)；large language model(abstract)；language model(abstract)；post-training(abstract)

Comments Accepted at EMNLP Findings 2025

Traveling Salesman-Based Token Ordering Improves Stability in Homomorphically Encrypted Language Models

Donghwan Rho, Sieun Seo, Hyewon Sung, Chohong Min, Ernest K. Ryu

专题命中 效率与部署 ：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.LG

Comments 34 pages

Information Extraction from Conversation Transcripts: Neuro-Symbolic vs. LLM

Alice Saebom Kwak, Maria Alexeeva, Gus Hahn-Powell, Keith Alcock, Kevin McLaughlin, Doug McCorkle, Gabe McNunn, Mihai Surdeanu

机构 * Department of Linguistics, University of Arizona（亚利桑那大学语言学系） ； Lum AI ； Eocene Environmental Group（Eocene环境集团） 纠错

专题命中 效率与部署 ：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments 15 pages, 2 figures

From Rational Answers to Emotional Resonance: The Role of Controllable Emotion Generation in Language Models

Yurui Dong, Luozhijie Jin, Yao Yang, Bingjie Lu, Jiaxi Yang, Zhi Liu

机构 * State Key Laboratory on Technologies for Chinese Medicine Pharmaceutical Process Control and Intelligent Manufacture（中药制药过程控制与智能制造技术国家重点实验室） ； Zhejiang Lab（浙江实验室） 纠错

专题命中 效率与部署 ：language model(title,abstract)；LLM(abstract)；large language model(abstract)；分类 cs.CL

Comments 43 pages, 5 figures

ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference

Xiang Liu, Zhenheng Tang, Peijie Dong, Zeyu Li, Yue Liu, Bo Li, Xuming Hu, Xiaowen Chu

机构 * The Hong Kong University of Science and Technology (Guangzhou)（香港科学与技术大学（广州）） ； CSE, The Hong Kong University of Science and Technology（香港科学与技术大学计算机科学与工程系） ； Guangzhou HKUST Fok Ying Tung Research Institute（广州HKUST福 Ying Tung研究 institute） ； Terminus Technologies 纠错

专题命中 效率与部署 ：LLM(title,abstract)；large language model(abstract)；language model(abstract)；分类 cs.CL

Comments NeurIPS 2025

Robust ML-based Detection of Conventional, LLM-Generated, and Adversarial Phishing Emails Using Advanced Text Preprocessing

Deeksha Hareesha Kulal, Chidozie Princewill Arannonu, Afsah Anwar, Nidhi Rastogi, Quamar Niyaz

专题命中 效率与部署 ：LLM(title,abstract)；large language model(abstract)；language model(abstract)

Tokenization Disparities as Infrastructure Bias: How Subword Systems Create Inequities in LLM Access and Efficiency

Hailay Kidu Teklehaymanot, Wolfgang Nejdl

机构 * L3S Research Center（L3S研究所以） ； Leibniz University Hannover（莱比锡大学汉诺威分校） 纠错

专题命中 效率与部署 ：LLM(title)；large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments 6 pages 4 figures

MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamics

Bowei Guo, Shengkun Tang, Cong Zeng, Zhiqiang Shen

机构 * Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学） 纠错

专题命中 效率与部署 ：pretraining(title,abstract)；post-training(abstract)；分类 cs.LG

Comments International Conference on Computer Vision, ICCV 2025

ACCO: Accumulate While You Communicate for Communication-Overlapped Sharded LLM Training

Adel Nabli, Louis Fournier, Pierre Erbacher, Louis Serrano, Eugene Belilovsky, Edouard Oyallon

专题命中 效率与部署 ：LLM(title,abstract)；分类 cs.AI、cs.LG

Journal ref The Thirty-ninth Annual Conference on Neural Information Processing Systems, Dec 2025, San diego (Californie), United States

BridgeVLA: Input-Output Alignment for Efficient 3D Manipulation Learning with Vision-Language Models

Peiyan Li, Yixiang Chen, Hongtao Wu, Xiao Ma, Xiangnan Wu, Yan Huang, Liang Wang, Tao Kong, Tieniu Tan

机构 * CASIA（中国科学院自动化研究所） ； ByteDance Seed（字节跳动种子实验室） ； UCAS（中国科学院大学） ； FiveAges ； NJU（南京大学） 纠错

专题命中 效率与部署 ：language model(title,abstract)；分类 cs.AI

Comments NeurIPS 2025

State Space Prompting via Gathering and Spreading Spatio-Temporal Information for Video Understanding

Jiahuan Zhou, Kai Zhu, Zhenyu Cui, Zichen Liu, Xu Zou, Gang Hua

机构 * Wangxuan Institute of Computer Technology, Peking University（北京大学王轩计算机技术研究所） ； the Huazhong University of Science and Technology（华中科技大学） ； Amazon.com, Inc（亚马逊公司） 纠错

专题命中 效率与部署 ：prompting(title,abstract)

ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization

Zechun Liu, Changsheng Zhao, Hanxian Huang, Sijia Chen, Jing Zhang, Jiawei Zhao, Scott Roy, Lisa Jin, Yunyang Xiong, Yangyang Shi, Lin Xiao, Yuandong Tian, Bilge Soran, Raghuraman Krishnamoorthi, Tijmen Blankevoort, Vikas Chandra

机构 * Meta AI 纠错

专题命中 效率与部署 ：LLM(title)；分类 cs.CL、cs.AI、cs.LG

Comments NeurIPS 2025. Model weights are available at https://huggingface.co/collections/facebook/mobilellm-6722be18cb86c20ebe113e95

CTRL-Rec: Controlling Recommender Systems With Natural Language

Micah Carroll, Adeline Foote, Kevin Feng, Marcus Williams, Anca Dragan, W. Bradley Knox, Smitha Milli

机构 * MATS ； University of Washington（华盛顿大学） ； UC Berkeley（加州大学伯克利分校） ； UT Austin（得克萨斯大学奥斯汀分校） ； FAIR at Meta（Meta的FAIR） 纠错

专题命中 效率与部署 ：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

EmboMatrix: A Scalable Training-Ground for Embodied Decision-Making

Zixing Lei, Sheng Yin, Yichen Xiong, Yuanzhuo Ding, Wenhao Huang, Yuxi Wei, Qingyao Xu, Yiming Li, Weixin Li, Yunhong Wang, Siheng Chen

机构 * Shanghai Jiao Tong University（上海交通大学） ； Zhongguancun Academy（中关村学院） ； New York University（纽约大学） 纠错

专题命中 效率与部署 ：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.AI

Comments 10 pages 8 figures

K-ASTRO: Structure-Aware Adaptation of LLMs for Code Vulnerability Detection

Yifan Zhang, Michael Sandborn, Stefan Larson, Yu Huang, Kevin Leach

机构 * Vanderbilt University（范德比大学） 纠错

专题命中 效率与部署 ：LLM(abstract)；large language model(abstract)；language model(abstract)；分类 cs.LG

MoEs Are Stronger than You Think: Hyper-Parallel Inference Scaling with RoE

Soheil Zibakhsh, Mohammad Samragh, Kumari Nishu, Lauren Hannah, Arnav Kundu, Minsik Cho

机构 * Apple（苹果公司） ； University of California San Diego（加州大学圣地亚哥分校） 纠错

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI、cs.LG

Comments Corrected typo in arxiv abstract

SMEC: Rethinking Matryoshka Representation Learning for Retrieval Embedding Compression

Biao Zhang, Lixin Chen, Tong Liu, Bo Zheng

机构 * Taobao & Tmall Group of Alibaba（淘宝与天猫集团） 纠错

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments Accepted by EMNLP2025

Direct Multi-Token Decoding

Xuan Luo, Weizhi Wang, Xifeng Yan

机构 * Department of Computer Science, UC Santa Barbara（计算机科学系，加州大学圣芭芭拉分校） 纠错

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

SCAN: Self-Denoising Monte Carlo Annotation for Robust Process Reward Learning

Yuyang Ding, Xinyu Shi, Juntao Li, Xiaobo Liang, Zhaopeng Tu, Min Zhang

机构 * Soochow University（苏州大学） ； Tencent（腾讯） 纠错

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments NeurIPS 2025. Project page: https://scan-prm.github.io/

Highlighting What Matters: Promptable Embeddings for Attribute-Focused Image Retrieval

Siting Li, Xiang Gao, Simon Shaolei Du

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments NeurIPS 2025; 27 pages, 6 figures

Multitask finetuning and acceleration of chemical pretrained models for small molecule drug property prediction

Matthew Adrian, Yunsie Chung, Kevin Boyd, Saee Paliwal, Srimukh Prasad Veccham, Alan C. Cheng

机构 * Modeling & Informatics（建模与信息学） ； Merck & Co., Inc.（默克公司） ； BioNeMo 纠错

专题命中 效率与部署 ：foundation model(abstract)；pretraining(abstract)；分类 cs.LG

Z0-Inf: Zeroth Order Approximation for Data Influence

Narine Kokhlikyan, Kamalika Chaudhuri, Saeed Mahloujifar

机构 * FAIR at Meta（Meta 的 FAIR） 纠错

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)；分类 cs.LG

An Empirical Study for Representations of Videos in Video Question Answering via MLLMs

Zhi Li, Yanan Wang, Hao Niu, Julio Vizcarra, Masato Taya

专题命中 效率与部署 ：large language model(abstract)；language model(abstract)

Comments 6 pages, 3 figures

OmniDraft: A Cross-vocabulary, Online Adaptive Drafter for On-device Speculative Decoding

Ramchalam Kinattinkara Ramakrishnan, Zhaocong Yuan, Shaojie Zhuo, Chen Feng, Yicheng Lin, Chenzheng Su, Xiaopeng Zhang

机构 * Qualcomm AI Research（高通人工智能研究） 纠错

专题命中 效率与部署 ：LLM(abstract)；分类 cs.CL、cs.LG

Comments Accepted to NeurIPS 2025

NinA: Normalizing Flows in Action. Training VLA Models with Normalizing Flows

Denis Tarasov, Alexander Nikulin, Ilya Zisman, Albina Klepach, Nikita Lyubaykin, Andrei Polubarov, Alexander Derevyagin, Vladislav Kurenkov

机构 * ETH Zürich（苏黎世联邦理工学院） ； MIPT（莫斯科国立信息安全大学） ； Skoltech（斯克里普斯基技术大学） ； Innopolis University（因诺波利斯大学） ； HSE（俄罗斯高等经济学院） 纠错

专题命中 效率与部署 ：language model(abstract)；分类 cs.AI、cs.LG

Comments https://github.com/dunnolab/NinA/