Reading list
Links 20
Score: 0.8653771783355902
User feedback: None
Out links: 199298 Raw text: 199298https://gwern.net/doc/www/arxiv.org/16217a265b539d6d8256a7c3722eff1b2291bcd6.pdf
Training Larger Networks for Deep Reinforcement Learning Kei Ota 1 2 Devesh K. Jha 3 Asako Kanezaki 2 arXiv:2102.07920v1 [cs.LG] 16 Feb 2021 Abstract The success of deep learning in the computer vision and natural language processing communities can be attributed to training of very deep neural n...
Score: 0.8446923041463351
User feedback: None
Out links: 22741 Raw text: 22741https://arxiv.org/pdf/2109.00698.pdf
An Empirical Exploration in Quality Filtering of Text Data Leo Gao EleutherAI [email protected] arXiv:2109.00698v2 [cs.CL] 6 Oct 2021 Abstract While conventional wisdom suggests that more aggressively filtering data from lowquality sources like Common Crawl always monotonically improves the quality o...
Score: 0.8200421290709352
User feedback: None
Out links: 22698 Raw text: 22698https://arxiv.org/pdf/2205.10487.pdf
Scaling Laws and Interpretability of Learning from Repeated Data arXiv:2205.10487v1 [cs.LG] 21 May 2022 Danny Hernandez∗ Tom Brown, Tom Conerly, Nova DasSarma, Dawn Drain, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Tom Henighan, Tristan Hume, Scott Johnston, Ben Mann, Chris Olah, Catherine...
Score: 0.8138346215978637
User feedback: None
Out links: 2124090 Raw text: 2124090http://www.cs.toronto.edu/~hinton/absps/similarity.pdf
Similarity of Neural Network Representations Revisited Simon Kornblith 1 Mohammad Norouzi 1 Honglak Lee 1 Geoffrey Hinton 1 arXiv:1905.00414v4 [cs.LG] 19 Jul 2019 Abstract Recent work has sought to understand the behavior of neural networks by comparing representations between layers and between ...
Score: 0.8095551863742576
User feedback: None
Out links: 27661 Raw text: 27661https://arxiv.org/pdf/1912.01991.pdf
Self-Supervised Learning of Pretext-Invariant Representations Ishan Misra Laurens van der Maaten Facebook AI Research arXiv:1912.01991v1 [cs.CV] 4 Dec 2019 Abstract Pretext Image Transform The goal of self-supervised learning from images is to construct image representations that are semanticall...
Score: 0.6955710721763025
User feedback: None
Out links: 3641987 Raw text: 3641987https://arxiv.org/pdf/2412.06769
Training Large Language Models to Reason in a Continuous Latent Space arXiv:2412.06769v1 [cs.CL] 9 Dec 2024 Shibo Hao1,2,∗ , Sainbayar Sukhbaatar1 , DiJia Su1 , Xian Li1 , Zhiting Hu2 , Jason Weston1 , Yuandong Tian1 1 FAIR at Meta, 2 UC San Diego ∗ Work done at Meta Large language models (LLM...
Score: 0.6955710721763025
User feedback: None
Out links: 4724552 Raw text: 4724552https://arxiv.org/pdf/2412.06769v1
Training Large Language Models to Reason in a Continuous Latent Space arXiv:2412.06769v1 [cs.CL] 9 Dec 2024 Shibo Hao1,2,∗ , Sainbayar Sukhbaatar1 , DiJia Su1 , Xian Li1 , Zhiting Hu2 , Jason Weston1 , Yuandong Tian1 1 FAIR at Meta, 2 UC San Diego ∗ Work done at Meta Large language models (LLM...
Score: 0.665546287731564
User feedback: None
Out links: 4724551 Raw text: 4724551https://www.microsoft.com/en-us/research/uploads/prod/2024/12/P4TechReport.pdf
Phi-4 Technical Report Marah Abdin Ronen Eldan Mojan Javaheripi Yuanzhi Li Eric Price Shital Shah Dingli Yu Jyoti Aneja Suriya Gunasekar Piero Kauffmann Weishung Liu Gustavo de Rosa Xin Wang Cyril Zhang Harkirat Behl Michael Harrison James R. Lee Caio C. T. Mendes Olli Saarikivi Rachel Ward Yi Zha...
Score: 0.6625596901056974
User feedback: None
Out links: 5666106 Raw text: 5666106https://arxiv.org/pdf/2410.22884
Stealing User Prompts from Mixture of Experts Itay Yona1 , Ilia Shumailov1 , Jamie Hayes1 and Nicholas Carlini1 Mixture-of-Experts (MoE) models improve the efficiency and scalability of dense language models by routing each token to a small number of experts in each layer. In this paper, we show ho...
Score: 0.6445696493191402
User feedback: None
Out links: 3601338 Raw text: 3601338https://arxiv.org/pdf/2412.06264
Flow Matching Guide and Code Yaron Lipman1 , Marton Havasi1 , Peter Holderrieth2 , Neta Shaul3 , Matt Le1 , Brian Karrer1 , Ricky T. Q. Chen1 , David Lopez-Paz1 , Heli Ben-Hamu3 , Itai Gat1 arXiv:2412.06264v1 [cs.LG] 9 Dec 2024 1 FAIR at Meta, 2 MIT CSAIL, 3 Weizmann Institute of Science Flow Ma...
Score: 0.6137844130486095
User feedback: None
Out links: 4487463 Raw text: 4487463https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/LearningBM25MSRTechReport.pdf
A Machine Learning Approach for Improved BM25 Retrieval Krysta M. Svore and Christopher J. C. Burges Microsoft Research One Microsoft Way Redmond, WA 98052 {ksvore,cburges}@microsoft.com Microsoft Research Technical Report MSR-TR-2009-92 July 30, 2009 Abstract BM25 is one of the most widely used in...
Score: 0.6056453874616692
User feedback: None
Out links: 5323804 Raw text: 5323804Deliberative Alignment: Reasoning Enables Safer Language Models Melody Y. Guan∗ Manas Joglekar Eric Wallace Alec Heylar Rachel Dias Andrea Vallone Hyung Won Chung Sam Toyer Johannes Heidecke Saachi Jain Hongyu Ren Alex Beutel Boaz Barak Jason Wei Amelia Glaese OpenAI Abstract As large-scale lang...
Score: 0.5485596078255999
User feedback: None
Out links: 5351856 Raw text: 5351856https://arxiv.org/pdf/2412.15042
Compiling C to Safe Rust, Formalized arXiv:2412.15042v1 [cs.PL] 19 Dec 2024 AYMERIC FROMHERZ, Inria, France JONATHAN PROTZENKO, Microsoft Azure Research, USA The popularity of the Rust language continues to explode; yet, many critical codebases remain authored in C, and cannot be realistically rew...
Score: 0.5259696888416368
User feedback: None
Out links: 17967 Raw text: 17967https://drwho.virtadpt.net/files/mov.pdf
mov is Turing-complete Stephen Dolan Computer Laboratory, University of Cambridge [email protected] Abstract registers for now, but later we show how their number can be reduced without losing expressiveness. We have the following instructions (if you like RISC) or addressing modes (if yo...
Score: 0.19811670196212236
User feedback: None
Out links: 198441 Raw text: 198441https://arxiv.org/pdf/2411.10958
...
Score: 0.19811670196212236
User feedback: None
Out links: 857603 Raw text: 857603https://arxiv.org/pdf/2411.14402
...
Score: 0.19811670196212236
User feedback: None
Out links: 567888 Raw text: 567888https://arxiv.org/pdf/2411.13676
...
Score: 0.19811670196212236
User feedback: None
Out links: 911729 Raw text: 911729https://www2.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-1.pdf
...
Score: 0.19811670196212236
User feedback: None
Out links: 988322 Raw text: 988322https://arxiv.org/pdf/2411.15131
...
Score: 0.19811670196212236
User feedback: None
Out links: 15364 Raw text: 15364https://arxiv.org/pdf/2408.00724
...