Open library
The essential reading of generative AI and agents, open in a clean reading view, with Meow and Standard summaries when available. No account needed to read.
560 papers · 558 with a Meow summary · 333 posts
The selected papers and blogs: the essential papers of generative AI and agents, and the key lab publications and essays, each with Meow and Standard summaries.
Albert Gu, Tri Dao
Foundation models, now powering most of the exciting applications in deep learning, are almost universally based on the Transformer architecture and its core attention module. Many subquadratic-time architectures such as linear attention, gated convolution and recurrent models, and structured state space models (SSMs)…
Tri Dao, Albert Gu
While Transformers have been the main architecture behind deep learning’s success in language modeling, state-space models (SSMs) such as Mamba have recently been shown to match or outperform Transformers at small to medium scale. We show that these families of models are actually quite closely related, and develop a…
Albert Gu, Karan Goel, Christopher Ré
A central goal of sequence modeling is designing a single principled model that can address sequence data across a range of modalities and tasks, particularly on long-range dependencies. Although conventional models including RNNs, CNNs, and Transformers have specialized variants for capturing long dependencies, they…
Bo Peng Eric Alcaide Quentin Anthony, Alon Albalak Samuel Arcadinho Stella Biderman Huanqi Cao Xin Cheng, Michael Chung Xingjian Du Matteo Grella Kranthi Kiran GV Xuzheng He et al.
Transformers have revolutionized almost all natural language processing (NLP) tasks but suffer from memory and computational complexity that scales quadratically with sequence length. In contrast, recurrent neural networks (RNNs) exhibit linear scaling in memory and computational requirements but struggle to match the…
Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pappas et al.
Transformers achieve remarkable performance in several tasks but due to their quadratic complexity, with respect to the input’s length, they are prohibitively slow for very long sequences. To address this limitation, we express the self-attention as a linear dot-product of kernel feature maps and make use of the assoc…
Ofir Press, Noah A. Smith, Mike Lewis
Since the introduction of the transformer model by [Vaswani et al. (2017)](#bib.bib39), a fundamental question has yet to be answered: how does a model achieve extrapolation at inference time for sequences that are longer than it saw during training? We first show that extrapolation can be enabled by simply changing t…
Bowen Peng, Jeffrey Quesnelle, Honglu Fan et al.
Rotary Position Embeddings (RoPE) have been shown to effectively encode positional information in transformer-based language models. However, these models fail to generalize past the sequence length they were trained on. We present YaRN (Yet another RoPE extensioN method), a compute-efficient method to extend the cont…
Guangxuan Xiao, Yuandong Tian, Beidi Chen et al.
Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challenges. Firstly, during the decoding stage, caching previous tokens’ Key and Value states (KV) consumes extensive memory. Secondly, popular LLM…
Jingyang Yuan, Huazuo Gao, Damai Dai, Junyu Luo, Liang Zhao, Zhengyan Zhang, Zhenda Xie, Y. X. Wei, Lean Wang, Zhiping Xiao, Yuqing Wang, Chong Ruan, Ming Zhang, Wenfeng Liang, Wangding Zeng
Long-context modeling is crucial for next-generation language models, yet the high computational cost of standard attention mechanisms poses significant computational challenges. Sparse attention offers a promising direction for improving efficiency while maintaining model capabilities. We present NSA, a Natively trai…
Songlin Yang, Jan Kautz, Ali Hatamizadeh
Linear Transformers have gained attention as efficient alternatives to standard Transformers, but their performance in retrieval and long-context tasks has been limited. To address these limitations, recent work has explored two distinct mechanisms: gating for adaptive memory control and the delta update rule for prec…
Cheng-Ping Hsieh, Simeng Sun, Samuel Kriman et al.
The needle-in-a-haystack (NIAH) test, which examines the ability to retrieve a piece of information (the “needle”) from long distractor texts (the “haystack”), has been widely adopted to evaluate long-context language models (LMs). However, this simple retrieval-based test is indicative of only a superficial form of l…
Nelson F. Liu Kevin Lin John Hewitt Ashwin ParanjapeMichele Bevilacqua Fabio Petroni Percy LiangStanford University University of California, Berkeley Samaya AI
While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of language models on two tasks that require identifying relevant information in their input contexts: multi-document question answering and key-va…
Google Research
Nested Learning argues that model architecture and training rules are levels of one learning system, a shift intended to help models learn new tasks without sacrificing old skills.
Qwen Team · Qwen
Qwen’s first open-source 7B and 14B Instruct checkpoints target 1M-token contexts, backed by an inference framework built for long inputs.
Lilian Weng
When a Transformer needs more context, less memory, or faster inference, which part of attention should change?
Sebastian Raschka, PhD · Sebastian Raschka
gpt-oss keeps the decoder-only Transformer but combines sparse expert routing, attention changes, and quantization to make large models more practical to run. The author’s benchmark comparisons are promising, but real-world reliability remains uncertain.
Simon Willison
Jev shifts an LLM from writing prose to returning typed decision scores. That suits classification, but its opaque numbers make evaluation and bias checks essential.
Andrej Karpathy
microgpt makes the GPT training-and-generation algorithm inspectable in 200 lines of dependency-free Python; its tiny scale is a feature for learning, not a model of production infrastructure.
Sebastian Raschka, PhD · Sebastian Raschka
The article’s useful map is about where each design saves work: cached state, attended positions, or full-attention layers.
Typesafe
TypeSafe’s thesis is that automating software calls for a different interface than chat: Jev returns typed, probabilistic decisions instead of free-form text.