{"id":4156,"date":"2024-07-22T11:45:09","date_gmt":"2024-07-22T03:45:09","guid":{"rendered":"https:\/\/nullthought.net\/?p=4156"},"modified":"2025-02-23T11:09:47","modified_gmt":"2025-02-23T03:09:47","slug":"%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a02024%e5%b9%b4%e5%ba%a6%e5%9b%bd%e9%99%85%e5%a4%a7%e4%bc%9a","status":"publish","type":"post","link":"https:\/\/nullthought.net\/?p=4156","title":{"rendered":"\u673a\u5668\u5b66\u4e602024\u5e74\u5ea6\u56fd\u9645\u5927\u4f1a"},"content":{"rendered":"\n

\u7b2c41\u754c\u673a\u5668\u5b66\u4e60\u56fd\u9645\u5927\u4f1a\uff08International Conference on Machine Learning, ICML 2024<\/a>\uff09\u4e8e2024\u5e747\u670821\u65e5\u523027\u65e5\u5728\u5965\u5730\u5229\u7ef4\u4e5f\u7eb3\u53ec\u5f00\u3002\u770b\u770b\u5927\u4f1a\u6240\u63a5\u53d7\u8bba\u6587\uff08Accepted papers\uff09\u5217\u8868<\/a>\uff0c\u53ef\u4ee5\u4e86\u89e3\u5f53\u524d\u673a\u5668\u5b66\u4e60\u7814\u7a76\u7684\u6700\u65b0\u52a8\u6001\u548c\u524d\u6cbf\u65b9\u5411<\/strong>\u3002<\/p>\n\n\n\n

\u8bba\u6587\u5217\u8868\u5982\u4e0b\uff08\u4e2d\u6587\u7531ChatGPT 4o\u7ffb\u8bd1\uff09\uff1a<\/p>\n\n\n\n

Perturb-and-Project: Differentially Private Similarities and Marginals<\/strong>
\u6270\u52a8\u4e0e\u6295\u5f71\uff1a\u5dee\u5206\u9690\u79c1\u76f8\u4f3c\u6027\u548c\u8fb9\u7f18\u5206\u5e03
\u4f5c\u8005\uff1aVincent Cohen-Addad, Tommaso d’Orsi, Alessandro Epasto, Vahab Mirrokni, Peilin Zhong<\/p>\n\n\n\n

Replicable Learning of Large-Margin Halfspaces<\/strong>
\u5927\u8fb9\u8ddd\u534a\u7a7a\u95f4\u7684\u53ef\u590d\u5236\u5b66\u4e60
\u4f5c\u8005\uff1aAlkis Kalavasis, Amin Karbasi, Kasper Green Larsen, Grigoris Velegkas, Felix Zhou<\/p>\n\n\n\n

Decoding-time Realignment of Language Models<\/strong>
\u8bed\u8a00\u6a21\u578b\u89e3\u7801\u65f6\u7684\u91cd\u65b0\u5bf9\u9f50
\u4f5c\u8005\uff1aTianlin Liu, Shangmin Guo, Leonardo Bianco*, Daniele Calandriello, Quentin Berthet, Felipe Llinares-L\u00f3pez, Jessica Hoffmann, Lucas Dixon, Michal Valko, Mathieu Blondel<\/p>\n\n\n\n

Target Networks and Over-parameterization Stabilize Off-policy Bootstrapping with Function Approximation<\/strong>
\u76ee\u6807\u7f51\u7edc\u548c\u8fc7\u5ea6\u53c2\u6570\u5316\u901a\u8fc7\u51fd\u6570\u903c\u8fd1\u7a33\u5b9a\u79bb\u7ebf\u7b56\u7565\u5f15\u5bfc
\u4f5c\u8005\uff1aFengdi Che, Chenjun Xiao, Jincheng Mei, Bo Dai, Ramki Gummadi, Oscar A Ramirez*, Christopher K Harris*, A. Rupam Mahmood, Dale Schuurmans<\/p>\n\n\n\n

Dynamic Correlation Clustering in Sublinear Update Time<\/strong>
\u6b21\u7ebf\u6027\u66f4\u65b0\u65f6\u95f4\u7684\u52a8\u6001\u76f8\u5173\u805a\u7c7b
\u4f5c\u8005\uff1aVincent Cohen-Addad, Silvio Lattanzi, Andreas Maggiori, Nikos Parotsidis<\/p>\n\n\n\n

PriorBoost: An Adaptive Algorithm for Learning from Aggregate Responses<\/strong>
PriorBoost\uff1a\u4e00\u79cd\u81ea\u9002\u5e94\u7b97\u6cd5\uff0c\u7528\u4e8e\u4ece\u6c47\u603b\u54cd\u5e94\u4e2d\u5b66\u4e60
\u4f5c\u8005\uff1aAdel Javanmard, Matthew Fahrbach, Vahab Mirrokni<\/p>\n\n\n\n

How Free is Parameter-Free Stochastic Optimization?<\/strong>
\u65e0\u53c2\u6570\u968f\u673a\u4f18\u5316\u6709\u591a\u81ea\u7531\uff1f
\u4f5c\u8005\uff1aAmit Attia, Tomer Koren<\/p>\n\n\n\n

Practical Performance Guarantees for Pipelined DNN Inference<\/strong>
\u6d41\u6c34\u7ebfDNN\u63a8\u7406\u7684\u5b9e\u7528\u6027\u80fd\u4fdd\u8bc1
\u4f5c\u8005\uff1aAaron Archer, Matthew Fahrbach, Kuikui Liu, Prakash Prabhu<\/p>\n\n\n\n

Regression with Multi-Expert Deferral<\/strong>
\u591a\u4e13\u5bb6\u63a8\u8fdf\u7684\u56de\u5f52
\u4f5c\u8005\uff1aAnqi Mao, Mehryar Mohri, Yutao Zhong<\/p>\n\n\n\n

Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond<\/strong>
\u901a\u8fc7\u57fa\u4e8e\u805a\u7c7b\u7684\u654f\u611f\u6027\u91c7\u6837\u5b9e\u73b0\u6570\u636e\u9ad8\u6548\u5b66\u4e60\uff1a\u57fa\u7840\u6a21\u578b\u53ca\u5176\u5e94\u7528
\u4f5c\u8005\uff1aKyriakos Axiotis, Vincent Cohen-Addad, Monika Henzinger, Sammy Jerome, Vahab Mirrokni, David Saulpic, David Woodruff, Michael Wunder<\/p>\n\n\n\n

Isometric Representation Learning for Disentangled Latent Space of Diffusion Models<\/strong>
\u7528\u4e8e\u89e3\u8026\u6f5c\u5728\u7a7a\u95f4\u7684\u7b49\u8ddd\u8868\u793a\u5b66\u4e60
\u4f5c\u8005\uff1aJaehoon Hahm, Junho Lee, Sunghyun Kim, Joonseok Lee<\/p>\n\n\n\n

Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs<\/strong>
\u4ece\u5b66\u751f\u4e2d\u5b66\u4e60\uff1a\u5e94\u7528t\u5206\u5e03\u63a2\u7d22\u51c6\u786e\u548c\u9ad8\u6548\u7684LLM\u683c\u5f0f
\u4f5c\u8005\uff1aJordan Dotzel, Yuzong Chen, Bahaa Kotb, Sushma Prasad, Gang Wu, Sheng Li, Mohamed S. Abdelfattah, Zhiru Zhang<\/p>\n\n\n\n

LEVI: Generalizable Fine-tuning via Layer-wise Ensemble of Different Views<\/strong>
LEVI\uff1a\u901a\u8fc7\u4e0d\u540c\u89c6\u89d2\u7684\u5206\u5c42\u96c6\u6210\u5b9e\u73b0\u53ef\u6cdb\u5316\u5fae\u8c03
\u4f5c\u8005\uff1aYuji Roh, Qingyun Liu, Huan Gui, Zhe Yuan, Yujin Tang, Steven Euijong Whang, Liang Liu, Shuchao Bi, Lichan Hong, Ed H. Chi, Zhe Zhao<\/p>\n\n\n\n

Out of the Ordinary: Spectrally Adapting Regression for Covariate Shift<\/strong>
\u975e\u5e38\u89c4\uff1a\u4e3a\u534f\u53d8\u91cf\u6f02\u79fb\u5149\u8c31\u8c03\u6574\u56de\u5f52
\u4f5c\u8005\uff1aBenjamin Eyre, Elliot Creager, David Madras, Vardan Papyan, Richard Zemel<\/p>\n\n\n\n

Privacy-Preserving Instructions for Aligning Large Language Models<\/strong>
\u9690\u79c1\u4fdd\u62a4\u7684\u6307\u4ee4\u7528\u4e8e\u5bf9\u9f50\u5927\u578b\u8bed\u8a00\u6a21\u578b
\u4f5c\u8005\uff1aDa Yu*, Peter Kairouz, Sewoong Oh, Zheng Xu<\/p>\n\n\n\n

Representation Surgery: Theory and Practice of Affine Steering<\/strong>
\u8868\u793a\u624b\u672f\uff1a\u4eff\u5c04\u8f6c\u5411\u7684\u7406\u8bba\u4e0e\u5b9e\u8df5
\u4f5c\u8005\uff1aShashwat Singh, Shauli Ravfogel*, Jonathan Herzig, Roee Aharoni, Ryan Cotterell, Ponnurangam Kumaraguru<\/p>\n\n\n\n

A Statistical Framework for Data-dependent Retrieval-Augmented Models<\/strong>
\u6570\u636e\u4f9d\u8d56\u7684\u68c0\u7d22\u589e\u5f3a\u6a21\u578b\u7684\u7edf\u8ba1\u6846\u67b6
\u4f5c\u8005\uff1aSoumya Basu, Ankit Singh Rawat, Manzil Zaheer<\/p>\n\n\n\n

Two Heads are Actually Better than One: Towards Better Adversarial Robustness via Transduction and Rejection<\/strong>
\u4e24\u4e2a\u5934\u786e\u5b9e\u6bd4\u4e00\u4e2a\u597d\uff1a\u901a\u8fc7\u8f6c\u5bfc\u548c\u62d2\u7edd\u5b9e\u73b0\u66f4\u597d\u7684\u5bf9\u6297\u9c81\u68d2\u6027
\u4f5c\u8005\uff1aNils Palumbo, Yang Guo, Xi Wu, Jiefeng Chen, Yingyu Liang, Somesh Jha<\/p>\n\n\n\n

Bayesian Regret Minimization in Offline Bandits<\/strong>
\u79bb\u7ebf\u5e26\u4e2d\u8d1d\u53f6\u65af\u540e\u6094\u6700\u5c0f\u5316
\u4f5c\u8005\uff1aMarek Petrik, Guy Tennenholtz, Mohammad Ghavamzadeh<\/p>\n\n\n\n

Break the Sequential Dependency of LLM Inference Using Lookahead Decoding<\/strong>
\u4f7f\u7528\u524d\u77bb\u89e3\u7801\u6253\u7834LLM\u63a8\u7406\u7684\u987a\u5e8f\u4f9d\u8d56
\u4f5c\u8005\uff1aYichao Fu, Peter Bailis, Ion Stoica, Hao Zhang<\/p>\n\n\n\n

Do Large Code Models Understand Programming Concepts? Counterfactual Analysis for Code Predicates<\/strong>
\u5927\u578b\u4ee3\u7801\u6a21\u578b\u80fd\u7406\u89e3\u7f16\u7a0b\u6982\u5ff5\u5417\uff1f\u4ee3\u7801\u8c13\u8bcd\u7684\u53cd\u4e8b\u5b9e\u5206\u6790
\u4f5c\u8005\uff1aAshish Hooda*, Mihai Christodorescu, Miltiadis Allamanis, Aaron Wilson, Kassem Fawaz, Somesh Jha<\/p>\n\n\n\n

DySLIM: Dynamics Stable Learning by Invariant Measure for Chaotic Systems<\/strong>
DySLIM\uff1a\u901a\u8fc7\u4e0d\u53d8\u6d4b\u5ea6\u5b9e\u73b0\u6df7\u6c8c\u7cfb\u7edf\u7684\u52a8\u6001\u7a33\u5b9a\u5b66\u4e60
\u4f5c\u8005\uff1aYair Schiff, Zhong Yi Wan, Jeffrey B. Parker, Stephan Hoyer, Volodymyr Kuleshov, Fei Sha, Leonardo Zepeda-N\u00fa\u00f1ez<\/p>\n\n\n\n

A Field Guide for Pacing Budget and ROS Constraints<\/strong>
\u9884\u7b97\u548cROS\u7ea6\u675f\u7684\u8282\u594f\u6307\u5357
\u4f5c\u8005\uff1aSantiago R. Balseiro, Kshipra Bhawalkar, Zhe Feng, Haihao Lu, Vahab Mirrokni, Balasubramanian Sivan, Di Wang<\/p>\n\n\n\n

How Private is DP-SGD?<\/strong>
DP-SGD\u6709\u591a\u79c1\u5bc6\uff1f
\u4f5c\u8005\uff1aLynn Chua, Badih Ghazi, Pritish Kamath, Ravi Kumar, Pasin Manurangsi, Amer Sinha, Chiyuan Zhang<\/p>\n\n\n\n

Improved Differentially Private and Lazy Online Convex Optimization: Lower Regret without Smoothness Requirements<\/strong>
\u6539\u8fdb\u7684\u5dee\u5206\u9690\u79c1\u548c\u60f0\u6027\u5728\u7ebf\u51f8\u4f18\u5316\uff1a\u65e0\u9700\u5149\u6ed1\u6027\u7684\u4f4e\u540e\u6094
\u4f5c\u8005\uff1aNaman Agarwal, Satyen Kale, Karan Singh, Abhradeep Guha Thakurta<\/p>\n\n\n\n

LayerMerge: Neural Network Depth Compression through Layer Pruning and Merging<\/strong>
LayerMerge\uff1a\u901a\u8fc7\u5c42\u526a\u679d\u548c\u5408\u5e76\u5b9e\u73b0\u795e\u7ecf\u7f51\u7edc\u6df1\u5ea6\u538b\u7f29
\u4f5c\u8005\uff1aJinuk Kim, Marwa El Halabi, Mingi Ji, Hyun Oh Song<\/p>\n\n\n\n

Learning and Forgetting Unsafe Examples in Large Language Models<\/strong>
\u5927\u578b\u8bed\u8a00\u6a21\u578b\u4e2d\u7684\u5b66\u4e60\u4e0e\u9057\u5fd8\u4e0d\u5b89\u5168\u793a\u4f8b
\u4f5c\u8005\uff1aJiachen Zhao, Zhun Deng, David Madras, James Zou, Mengye Ren<\/p>\n\n\n\n

A Near-Linear Time Approximation Algorithm for Beyond-Worst-Case Graph Clustering<\/strong>
\u8d85\u8d8a\u6700\u574f\u60c5\u51b5\u56fe\u805a\u7c7b\u7684\u8fd1\u7ebf\u6027\u65f6\u95f4\u8fd1\u4f3c\u7b97\u6cd5
\u4f5c\u8005\uff1aVincent Cohen-Addad, Tommaso d’Orsi, Aida Mousavifar<\/p>\n\n\n\n

The Non-linear F-Design and Applications to Interactive Learning<\/strong>
\u975e\u7ebf\u6027F\u8bbe\u8ba1\u53ca\u5176\u5728\u4ea4\u4e92\u5b66\u4e60\u4e2d\u7684\u5e94\u7528
\u4f5c\u8005\uff1aAlekh Agarwal, Jian Qian, Alexander Rakhlin, Tong Zhang<\/p>\n\n\n\n

Pi-DUAL: Using Privileged Information to Distinguish Clean from Noisy Labels<\/strong>
Pi-DUAL\uff1a\u5229\u7528\u7279\u6743\u4fe1\u606f\u533a\u5206\u5e72\u51c0\u6807\u7b7e\u548c\u566a\u58f0\u6807\u7b7e
\u4f5c\u8005\uff1aKe Wang, Guillermo Ortiz-Jimenez, Rodolphe Jenatton, Mark Collier, Efi Kokiopoulou, Pascal Frossard<\/p>\n\n\n\n

Position: Cracking the Code of Cascading Disparity Towards Marginalized Communities<\/strong>
\u7834\u89e3\u8fb9\u7f18\u5316\u793e\u533a\u7ea7\u8054\u5dee\u8ddd\u7684\u5bc6\u7801
\u4f5c\u8005\uff1aGolnoosh Farnadi, Mohammad Havaei, Negar Rostamzadeh<\/p>\n\n\n\n

Unmasking Vulnerabilities: Cardinality Sketches Under Adaptive Inputs<\/strong>
\u63ed\u793a\u6f0f\u6d1e\uff1a\u81ea\u9002\u5e94\u8f93\u5165\u4e0b\u7684\u57fa\u6570\u8349\u56fe
\u4f5c\u8005\uff1aSara Ahmadian, Edith Cohen<\/p>\n\n\n\n

What is Dataset Distillation Learning?<\/strong>
\u4ec0\u4e48\u662f\u6570\u636e\u96c6\u84b8\u998f\u5b66\u4e60\uff1f
\u4f5c\u8005\uff1aWilliam Yang, Ye Zhu, Zhiwei Deng, Olga Russakovsky<\/p>\n\n\n\n

Can Looped Transformers Learn to Implement Multi-step Gradient Descent for In-context Learning?<\/strong>
\u5faa\u73af\u53d8\u538b\u5668\u80fd\u5426\u5b66\u4e60\u5b9e\u73b0\u4e0a\u4e0b\u6587\u5b66\u4e60\u7684\u591a\u6b65\u68af\u5ea6\u4e0b\u964d\uff1f
\u4f5c\u8005\uff1aKhashayar Gatmiry, Nikunj Saunshi, Sashank J. Reddi, Stefanie Jegelka, Sanjiv Kumar<\/p>\n\n\n\n

Cell2Sentence: Teaching Large Language Models the Language of Biology<\/strong>
Cell2Sentence\uff1a\u6559\u5927\u578b\u8bed\u8a00\u6a21\u578b\u751f\u7269\u5b66\u8bed\u8a00
\u4f5c\u8005\uff1aDaniel Levine, Syed A Rizvi, Sacha L\u00e9vy, Nazreen Pallikkavaliyaveetil, David Zhang, Xingyu Chen, Sina Ghadermarzi, Ruiming Wu, Zihe Zheng, Ivan Vrkic, Anna Zhong, Daphne Raskin, Insu Han, Antonio Henrique de Oliveira Fonseca, Josue Ortega Caro, Amin Karbasi, Rahul Madhav Dhodapkar, David van Dijk<\/p>\n\n\n\n

Consistent Submodular Maximization<\/strong>
\u4e00\u81f4\u6027\u5b50\u6a21\u6700\u5927\u5316
\u4f5c\u8005\uff1aPaul Duetting, Federico Fusco, Silvio Lattanzi, Ashkan Norouzi-Fard, Morteza Zaddimoghadam<\/p>\n\n\n\n

Controlled Decoding from Language Models<\/strong>
\u8bed\u8a00\u6a21\u578b\u7684\u53d7\u63a7\u89e3\u7801
\u4f5c\u8005\uff1aSidharth Mudgal, Jong Lee, Harish Ganapathy, YaGuang Li, Tao Wang*, Yanping Huang, Zhifeng Chen, Heng-Tze Cheng, Michael Collins, Trevor Strohman, Jilin Chen, Alex Beutel*, Ahmad Beirami<\/p>\n\n\n\n

Differentially Private Domain Adaptation with Theoretical Guarantees<\/strong>
\u5177\u6709\u7406\u8bba\u4fdd\u8bc1\u7684\u5dee\u5206\u9690\u79c1\u9886\u57df\u9002\u5e94
\u4f5c\u8005\uff1aRaef Bassily, Corinna Cortes, Anqi Mao, Mehryar Mohri<\/p>\n\n\n\n

Eluder-Based Regret for Stochastic Contextual MDPs<\/strong>
\u968f\u673a\u4e0a\u4e0b\u6587MDP\u7684Eluder\u578b\u540e\u6094
\u4f5c\u8005\uff1aOrin Levy, Asaf Cassel, Alon Cohen, Yishay Mansour<\/p>\n\n\n\n

A Minimaximalist Approach to Reinforcement Learning from Human Feedback<\/strong>
\u4ece\u4eba\u7c7b\u53cd\u9988\u4e2d\u5b66\u4e60\u7684\u6781\u5c0f\u6781\u5927\u65b9\u6cd5
\u4f5c\u8005\uff1aGokul Swamy*, Christoph Dann, Rahul Kidambi, Zhiwei Steven Wu, Alekh Agarwal<\/p>\n\n\n\n

Multi-View Stochastic Block Models<\/strong>
\u591a\u89c6\u89d2\u968f\u673a\u5757\u6a21\u578b
\u4f5c\u8005\uff1aVincent Cohen-Addad, Tommaso d’Orsi, Silvio Lattanzi, Rajai Nasser<\/p>\n\n\n\n

Near-Optimal Regret in Linear MDPs with Aggregate Bandit Feedback<\/strong>
\u5177\u6709\u805a\u5408\u5e26\u53cd\u9988\u7684\u7ebf\u6027MDP\u7684\u8fd1\u6700\u4f18\u540e\u6094
\u4f5c\u8005\uff1aAsaf Cassel, Haipeng Luo, Aviv Rosenberg, Dmitry Sotnikov<\/p>\n\n\n\n

Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models<\/strong>
Patchscopes\uff1a\u68c0\u67e5\u8bed\u8a00\u6a21\u578b\u9690\u85cf\u8868\u793a\u7684\u7edf\u4e00\u6846\u67b6
\u4f5c\u8005\uff1aAsma Ghandeharioun, Avi Caciularu, Adam Pearce, Lucas Dixon, Mor Geva<\/p>\n\n\n\n

Robust Inverse Graphics via Probabilistic Inference<\/strong>
\u901a\u8fc7\u6982\u7387\u63a8\u7406\u5b9e\u73b0\u9c81\u68d2\u9006\u5411\u56fe\u5f62\u5b66
\u4f5c\u8005\uff1aTuan Anh Le, Pavel Sountsov, Matthew Douglas Hoffman, Ben Lee, Brian Patton, Rif A. Saurous<\/p>\n\n\n\n

Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation<\/strong>
\u8bc4\u5206\u8eab\u4efd\u84b8\u998f\uff1a\u9884\u8bad\u7ec3\u6269\u6563\u6a21\u578b\u7684\u6307\u6570\u5feb\u901f\u84b8\u998f\u7528\u4e8e\u4e00\u6b65\u751f\u6210
\u4f5c\u8005\uff1aMingyuan Zhou, Huangjie Zheng, Zhendong Wang, Mingzhang Yin, Hai Huang<\/p>\n\n\n\n

Tandem Transformers for Inference Efficient LLMs<\/strong>
\u7528\u4e8e\u9ad8\u6548\u63a8\u7406\u7684\u4e32\u8054\u53d8\u538b\u5668
\u4f5c\u8005\uff1aAishwarya P S, Pranav Ajit Nair, Yashas Samaga B L, Toby James Boyd, Sanjiv Kumar, Prateek Jain, Praneeth Netrapalli<\/p>\n\n\n\n

Transforming and Combining Rewards for Aligning Large Language Models<\/strong>
\u8c03\u6574\u548c\u7ec4\u5408\u5956\u52b1\u4ee5\u5bf9\u9f50\u5927\u578b\u8bed\u8a00\u6a21\u578b
\u4f5c\u8005\uff1aZihao Wang, Chirag Nagpal, Jonathan Berant, Jacob Eisenstein, Alexander D’Amour, Sanmi Koyejo, Victor Veitch<\/p>\n\n\n\n

USTAD: Unified Single-Model Training Achieving Diverse Scores for Information Retrieval<\/strong>
USTAD\uff1a\u5b9e\u73b0\u4fe1\u606f\u68c0\u7d22\u591a\u6837\u5316\u5f97\u5206\u7684\u7edf\u4e00\u5355\u6a21\u578b\u8bad\u7ec3
\u4f5c\u8005\uff1aSeungyeon Kim, Ankit Singh Rawat, Manzil Zaheer, Wittawat Jitkrittum, Veeranjaneyulu Sadhanala, Sadeep Jayasumana, Aditya Krishna Menon, Rob Fergus, Sanjiv Kumar<\/p>\n\n\n\n

Adaptive Accompaniment with ReaLchords<\/strong>
\u4f7f\u7528ReaLchords\u8fdb\u884c\u81ea\u9002\u5e94\u4f34\u594f
\u4f5c\u8005\uff1aYusong Wu, Tim Cooijmans, Kyle Kastner, Adam Roberts, Ian Simon, Alexander Scarlatos, Chris Donahue, Cassie Tarakajian, Shayegan Omidshafiei*, Aaron Courville, Pablo Samuel Castro, Natasha Jaques, Cheng-Zhi Anna Huang<\/p>\n\n\n\n

A Decoder-Only Foundation Model for Time-Series Forecasting<\/strong>
\u7528\u4e8e\u65f6\u95f4\u5e8f\u5217\u9884\u6d4b\u7684\u4ec5\u89e3\u7801\u5668\u57fa\u7840\u6a21\u578b
\u4f5c\u8005\uff1aAbhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou<\/p>\n\n\n\n

Deep Fusion: Efficient Network Training via Pre-trained Initializations<\/strong>
\u6df1\u5ea6\u878d\u5408\uff1a\u901a\u8fc7\u9884\u8bad\u7ec3\u521d\u59cb\u5316\u5b9e\u73b0\u9ad8\u6548\u7f51\u7edc\u8bad\u7ec3
\u4f5c\u8005\uff1aHanna Mazzawi, Javier Gonzalvo, Michael Wunder, Sammy Jerome, Benoit Dherin<\/p>\n\n\n\n

Extracting Training Data from Document-Based VQA Models<\/strong>
\u4ece\u57fa\u4e8e\u6587\u6863\u7684\u89c6\u89c9\u95ee\u7b54\u6a21\u578b\u4e2d\u63d0\u53d6\u8bad\u7ec3\u6570\u636e
\u4f5c\u8005\uff1aFrancesco Pinto, Nathalie Rauschmayr, Florian Tramer, Philip Torr, Federico Tombari<\/p>\n\n\n\n

FrameQuant: Flexible Low-Bit Quantization for Transformers<\/strong>
FrameQuant\uff1a\u7528\u4e8e\u53d8\u538b\u5668\u7684\u7075\u6d3b\u4f4e\u6bd4\u7279\u91cf\u5316
\u4f5c\u8005\uff1aHarshavardhan Adepu, Zhanpeng Zeng, Li Zhang, Vikas Singh<\/p>\n\n\n\n

H-Consistency Guarantees for Regression<\/strong>
\u56de\u5f52\u7684H-\u4e00\u81f4\u6027\u4fdd\u8bc1
\u4f5c\u8005\uff1aAnqi Mao, Mehryar Mohri, Yutao Zhong<\/p>\n\n\n\n

Implicit Bias of Policy Gradient in Linear Quadratic Control: Extrapolation to Unseen Initial States<\/strong>
\u7ebf\u6027\u4e8c\u6b21\u63a7\u5236\u4e2d\u7b56\u7565\u68af\u5ea6\u7684\u9690\u5f0f\u504f\u5dee\uff1a\u5411\u672a\u89c1\u521d\u59cb\u72b6\u6001\u7684\u5916\u63a8
\u4f5c\u8005\uff1aNoam Razin, Yotam Alexander, Edo Cohen-Karlik, Raja Giryes, Amir Globerson, Nadav Cohen<\/p>\n\n\n\n

Interpretability Illusions in the Generalization of Simplified Models<\/strong>
\u7b80\u5316\u6a21\u578b\u63a8\u5e7f\u4e2d\u7684\u53ef\u89e3\u91ca\u6027\u5e7b\u89c9
\u4f5c\u8005\uff1aDan Friedman*, Andrew Kyle Lampinen, Lucas Dixon, Danqi Chen, Asma Ghandeharioun<\/p>\n\n\n\n

Large Language Models Can Automatically Engineer Features for Few-Shot Tabular Learning<\/strong>
\u5927\u578b\u8bed\u8a00\u6a21\u578b\u53ef\u4ee5\u81ea\u52a8\u8bbe\u8ba1\u7279\u5f81\u4ee5\u8fdb\u884c\u5c11\u6837\u672c\u8868\u683c\u5b66\u4e60
\u4f5c\u8005\uff1aSungwon Han*, Jinsung Yoon, Sercan O Arik, Tomas Pfister<\/p>\n\n\n\n

MC-GTA: Metric-Constrained Model-Based Clustering Using Goodness-of-Fit Tests with Autocorrelations<\/strong>
MC-GTA\uff1a\u4f7f\u7528\u62df\u5408\u4f18\u5ea6\u6d4b\u8bd5\u548c\u81ea\u76f8\u5173\u7684\u5ea6\u91cf\u7ea6\u675f\u6a21\u578b\u805a\u7c7b
\u4f5c\u8005\uff1aZhangyu Wang, Gengchen Mai, Krzysztof Janowicz, Ni Lao<\/p>\n\n\n\n

Mean Estimation in the Add-Remove Model of Differential Privacy<\/strong>
\u5dee\u5206\u9690\u79c1\u52a0\u79fb\u6a21\u578b\u4e2d\u7684\u5747\u503c\u4f30\u8ba1
\u4f5c\u8005\uff1aAlex Kulesza, Ananda Suresh, Yuyan Wang<\/p>\n\n\n\n

More Benefits of Being Distributional: Second-Order Bounds for Reinforcement Learning<\/strong>
\u5206\u5e03\u5316\u7684\u66f4\u591a\u597d\u5904\uff1a\u5f3a\u5316\u5b66\u4e60\u7684\u4e8c\u9636\u754c\u9650
\u4f5c\u8005\uff1aKaiwen Wang, Owen Oertell, Alekh Agarwal, Nathan Kallus, Wen Sun<\/p>\n\n\n\n

Online Learning with Bounded Recall<\/strong>
\u6709\u754c\u56de\u5fc6\u7684\u5728\u7ebf\u5b66\u4e60
\u4f5c\u8005\uff1aJon Schneider, Kiran Vodrahalli<\/p>\n\n\n\n

Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity<\/strong>
\u79bb\u7fa4\u503c\u52a0\u6743\u5c42\u6b21\u7a00\u758f\u6027\uff08OWL\uff09\uff1a\u4fee\u526aLLM\u5230\u9ad8\u7a00\u758f\u6027\u7684\u79d8\u5bc6\u8c03\u6599
\u4f5c\u8005\uff1aLu Yin, You Wu, Zhenyu Zhang, Cheng-Yu Hsieh, Yaqing Wang, Yiling Jia, Gen Li, Ajay Kumar Jaiswal, Mykola Pechenizkiy, Yi Liang, Michael Bendersky, Zhangyang Wang, Shiwei Liu<\/p>\n\n\n\n

Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines<\/strong>
\u751f\u6210\u906e\u63a9\u8bed\u8a00\u6a21\u578b\u7684\u627f\u8bfa\u548c\u9677\u9631\uff1a\u7406\u8bba\u6846\u67b6\u548c\u5b9e\u7528\u6307\u5357
\u4f5c\u8005\uff1aYuchen Li, Alexandre Kirchmeyer, Aashay Mehta, Yilong Qin, Boris Dadachev, Kishore Papineni, Sanjiv Kumar, Andrej Risteski<\/p>\n\n\n\n

SCoRe: Submodular Combinatorial Representation Learning<\/strong>
SCoRe\uff1a\u5b50\u6a21\u7ec4\u5408\u8868\u793a\u5b66\u4e60
\u4f5c\u8005\uff1aAnay Majee, Suraj Kothawade, Krishnateja Killamsetty, Rishabh K Iyer<\/p>\n\n\n\n

Simplicity Bias via Global Convergence of Sharpness Minimization<\/strong>
\u901a\u8fc7\u9510\u5ea6\u6700\u5c0f\u5316\u7684\u5168\u5c40\u6536\u655b\u5b9e\u73b0\u7684\u7b80\u5355\u6027\u504f\u89c1
\u4f5c\u8005\uff1aKhashayar Gatmiry, Zhiyuan Li, Sashank J. Reddi, Stefanie Jegelka<\/p>\n\n\n\n

Auto-Linear Phenomenon in Subsurface Imaging<\/strong>
\u5730\u4e0b\u6210\u50cf\u4e2d\u7684\u81ea\u52a8\u7ebf\u6027\u73b0\u8c61
\u4f5c\u8005\uff1aYinan Feng, Yinpeng Chen, Peng Jin, Shihang Feng, Youzuo Lin<\/p>\n\n\n\n

FRAPP\u00c9: A Group Fairness Framework for Post-Processing Everything<\/strong>
FRAPP\u00c9\uff1a\u4e00\u4e2a\u7528\u4e8e\u540e\u5904\u7406\u6240\u6709\u5185\u5bb9\u7684\u7fa4\u4f53\u516c\u5e73\u6846\u67b6
\u4f5c\u8005\uff1aAlexandru Tifrea*, Preethi Lahoti, Ben Packer, Yoni Halpern, Ahmad Beirami, Flavien Prost<\/p>\n\n\n\n

Individualized Privacy Accounting via Subsampling with Applications in Combinatorial Optimization<\/strong>
\u901a\u8fc7\u5b50\u91c7\u6837\u5b9e\u73b0\u7684\u4e2a\u6027\u5316\u9690\u79c1\u8ba1\u91cf\u53ca\u5176\u5728\u7ec4\u5408\u4f18\u5316\u4e2d\u7684\u5e94\u7528
\u4f5c\u8005\uff1aBadih Ghazi, Pritish Kamath, Ravi Kumar, Pasin Manurangsi, Adam Sealfon<\/p>\n\n\n\n

Online Speculative Decoding<\/strong>
\u5728\u7ebf\u6295\u673a\u89e3\u7801
\u4f5c\u8005\uff1aXiaoxuan Liu, Lanxiang Hu, Peter Bailis, Alvin Cheung, Zhijie Deng, Ion Stoica, Hao Zhang<\/p>\n\n\n\n

The Pitfalls of Next-Token Prediction<\/strong>
\u4e0b\u4e00\u4e2a\u4ee4\u724c\u9884\u6d4b\u7684\u9677\u9631
\u4f5c\u8005\uff1aGregor Bachmann, Vaishnavh Nagarajan<\/p>\n\n\n\n

PolySketchFormer: Fast Transformers via Sketching Polynomial Kernels<\/strong>
PolySketchFormer\uff1a\u901a\u8fc7\u7ed8\u5236\u591a\u9879\u5f0f\u6838\u5b9e\u73b0\u5feb\u901f\u53d8\u538b\u5668
\u4f5c\u8005\uff1aPraneeth Kacham, Vahab Mirrokni, Peilin Zhong<\/p>\n\n\n\n

Position: Social Environment Design Should be Further Developed for AI-based Policy-Making<\/strong>
\u89c2\u70b9\uff1a\u793e\u4f1a\u73af\u5883\u8bbe\u8ba1\u5e94\u8fdb\u4e00\u6b65\u53d1\u5c55\u4ee5\u652f\u6301\u57fa\u4e8eAI\u7684\u653f\u7b56\u5236\u5b9a
\u4f5c\u8005\uff1aEdwin Zhang, Sadie Zhao, Tonghan Wang, Safwan Hossain, Henry Gasztowtt, Stephan Zheng, David C. Parkes, Milind Tambe, Yiling Chen<\/p>\n\n\n\n

Prompt-Tuning Latent Diffusion Models for Inverse Problems<\/strong>
\u7528\u4e8e\u9006\u95ee\u9898\u7684\u63d0\u793a\u8c03\u6574\u6f5c\u5728\u6269\u6563\u6a21\u578b
\u4f5c\u8005\uff1aHyungjin Chung, Jong Chul Ye, Peyman Milanfar, Mauricio Delbracio<\/p>\n\n\n\n

VideoPrism: A Foundational Visual Encoder for Video Understanding<\/strong>
\u4e00\u4e2a\u7528\u4e8e\u89c6\u9891\u7406\u89e3\u7684\u57fa\u7840\u89c6\u89c9\u7f16\u7801\u5668
\u4f5c\u8005\uff1aLong Zhao, Nitesh Bharadwaj Gundavarapu, Liangzhe Yuan, Hao Zhou, Shen Yan, Jennifer J. Sun, Luke Friedman, Rui Qian, Tobias Weyand, Yue Zhao*, Rachel Hornung, Florian Schroff, Ming-Hsuan Yang, David A Ross, Huisheng Wang, Hartwig Adam, Mikhail Sirotenko, Ting Liu, Boqing Gong<\/p>\n\n\n\n

RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback<\/strong>
RLAIF vs. RLHF\uff1a\u901a\u8fc7AI\u53cd\u9988\u6269\u5c55\u4eba\u7c7b\u53cd\u9988\u5f3a\u5316\u5b66\u4e60
\u4f5c\u8005\uff1aHarrison Lee, Samrat Phatale, Hassan Mansoor, Thomas Mesnard, Johan Ferret, Kellie Ren Lu, Colton Bishop, Ethan Hall, Victor Carbune, Abhinav Rastogi, Sushant Prakash<\/p>\n\n\n\n

From Self-Attention to Markov Models: Unveiling the Dynamics of Generative Transformers<\/strong>
\u4ece\u81ea\u6ce8\u610f\u529b\u5230\u9a6c\u5c14\u53ef\u592b\u6a21\u578b\uff1a\u63ed\u793a\u751f\u6210\u53d8\u538b\u5668\u7684\u52a8\u6001
\u4f5c\u8005\uff1aMuhammed Emrullah Ildiz, Yixiao Huang, Yingcong Li, Ankit Singh Rawat, Samet Oymak<\/p>\n\n\n\n

Generalized Neural Collapse for a Large Number of Classes<\/strong>
\u5927\u91cf\u7c7b\u522b\u7684\u5e7f\u4e49\u795e\u7ecf\u574d\u584c
\u4f5c\u8005\uff1aJiachen Jiang, Jinxin Zhou, Peng Wang, Qing Qu, Dustin G. Mixon, Chong You, Zhihui Zhu<\/p>\n\n\n\n

High-Dimensional Geometric Streaming for Nearly Low Rank Data<\/strong>
\u8fd1\u4f4e\u79e9\u6570\u636e\u7684\u9ad8\u7ef4\u51e0\u4f55\u6d41
\u4f5c\u8005\uff1aHossein Esfandiari, Praneeth Kacham, Vahab Mirrokni, David Woodruff, Peilin Zhong<\/p>\n\n\n\n

Improved Communication-Privacy Trade-Offs in L2 Mean Estimation Under Streaming Differential Privacy<\/strong>
\u5728\u6d41\u5f0f\u5dee\u5206\u9690\u79c1\u4e0bL2\u5747\u503c\u4f30\u8ba1\u4e2d\u6539\u8fdb\u7684\u901a\u4fe1-\u9690\u79c1\u6743\u8861
\u4f5c\u8005\uff1aWei-Ning Chen, Berivan Isik, Peter Kairouz, Albert No, Sewoong Oh, Zheng Xu<\/p>\n\n\n\n

On Discrete Prompt Optimization for Diffusion Models<\/strong>
\u5173\u4e8e\u6269\u6563\u6a21\u578b\u7684\u79bb\u6563\u63d0\u793a\u4f18\u5316
\u4f5c\u8005\uff1aRuochen Wang, Ting Liu, Cho-Jui Hsieh, Boqing Gong<\/p>\n\n\n\n

OSSCAR: One-Shot Structured Pruning in Vision and Language Models with Combinatorial Optimization<\/strong>
OSSCAR\uff1a\u901a\u8fc7\u7ec4\u5408\u4f18\u5316\u5728\u89c6\u89c9\u548c\u8bed\u8a00\u6a21\u578b\u4e2d\u7684\u4e00\u6b21\u6027\u7ed3\u6784\u5316\u526a\u679d
\u4f5c\u8005\uff1aXiang Meng, Shibal Ibrahim, Kayhan Behdin, Hussein Hazimeh, Natalia Ponomareva, Rahul Mazumder<\/p>\n\n\n\n

Weisfeiler-Leman at the Margin: When More Expressivity Matters<\/strong>
Weisfeiler-Leman\u5728\u8fb9\u7f18\uff1a\u5f53\u66f4\u591a\u8868\u73b0\u529b\u5f88\u91cd\u8981\u65f6
\u4f5c\u8005\uff1aBilly Joe Franks, Christopher Morris, Ameya Velingker, Floris Geerts<\/p>\n","protected":false},"excerpt":{"rendered":"

\u7b2c41\u754c\u673a\u5668\u5b66\u4e60\u56fd\u9645\u5927\u4f1a\uff08International Conference on Machine Learni […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[8,36],"tags":[39,63,95,66],"class_list":{"0":"post-4156","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"hentry","6":"category-tech","7":"category-36","8":"tag-ai","10":"tag-transformer","11":"tag-google"},"rttpg_featured_image_url":null,"rttpg_author":{"display_name":"NullThought","author_link":"https:\/\/nullthought.net\/?author=1"},"rttpg_comment":0,"rttpg_category":"Tech<\/a> \u79d1\u5b66<\/a>","rttpg_excerpt":"\u7b2c41\u754c\u673a\u5668\u5b66\u4e60\u56fd\u9645\u5927\u4f1a\uff08International Conference on Machine Learni…","_links":{"self":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4156","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4156"}],"version-history":[{"count":2,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4156\/revisions"}],"predecessor-version":[{"id":4158,"href":"https:\/\/nullthought.net\/index.php?rest_route=\/wp\/v2\/posts\/4156\/revisions\/4158"}],"wp:attachment":[{"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4156"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4156"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nullthought.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4156"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}