Content-based routing succeeds only when models provide bidirectional context and perform pairwise comparisons, with bidirectional Mamba plus rank-1 projection reaching 99.7% precision at linear inference cost.
hub Canonical reference
RWKV: Reinventing RNNs for the Transformer Era
Canonical reference. 84% of citing Pith papers cite this work as background.
57
Pith papers citing it
Background
84% of classified citations
abstract
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 same performance as Transformers due to limitations in parallelization and scalability. We propose a novel model architecture, Receptance Weighted Key Value (RWKV), that combines the efficient parallelizable training of transformers with the efficient inference of RNNs. Our approach leverages a linear attention mechanism and allows us to formulate the model as either a Transformer or an RNN, thus parallelizing computations during training and maintains constant computational and memory complexity during inference. We scale our models as large as 14 billion parameters, by far the largest dense RNN ever trained, and find RWKV performs on par with similarly sized Transformers, suggesting future work can leverage this architecture to create more efficient models. This work presents a significant step towards reconciling trade-offs between computational efficiency and model performance in sequence processing tasks.
hub tools
citation-role summary
background 17
baseline 1
dataset 1
citation-polarity summary
representative citing papers
EQ-VMamba adds rotation-equivariant cross-scan and group Mamba blocks to enforce end-to-end rotation equivariance, yielding better rotation robustness, competitive accuracy, and roughly 50% fewer parameters than non-equivariant baselines across classification, segmentation, and super-resolution.
TTT layers treat the hidden state as a trainable model updated at test time, allowing linear-complexity sequence models to scale perplexity reduction with context length unlike Mamba.
Mamba is a linear-time sequence model using input-dependent selective SSMs that achieves SOTA results across modalities and matches twice-larger Transformers on language modeling with 5x higher inference throughput.
PSR-NQS makes recurrent neural quantum states scalable for variational Monte Carlo by using parallel scan recurrence, reaching accurate results on 52x52 two-dimensional lattices.
Adding temporal memory via LIF, precision-weighted gating, and anticipatory prediction to MoE routers recovers effective expert selection at distribution transitions, with ablation confirming a super-additive beta-ant interaction.
Sparse prefix caching via dynamic programming for optimal checkpoint placement under overlap distributions improves the Pareto frontier for recurrent and hybrid LLM serving on shared-prefix data.
Winner-take-all spiking self-attention replaces softmax in spiking transformers to support language modeling on 16 datasets with spike-driven, energy-efficient architectures.
ZipMap achieves linear-time bidirectional 3D reconstruction by zipping image collections into a compact stateful representation via test-time training layers.
DELTA partitions layers into full, delta, and sparse groups to select salient tokens via aggregated attention scores, matching full-attention accuracy on AIME and GPQA while cutting attended tokens up to 4.25x and achieving 1.54x speedup.
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
Transformers and SSMs are unified through structured state space duality, producing a 2-8X faster Mamba-2 model that remains competitive with Transformers.
DGNO parameterizes integral kernels with discontinuous Galerkin elements for heterogeneous defocus deblurring in pathology images and reports superior performance over prior methods.
A 194M-parameter spiking dual-path model trained on 3B Chinese-English tokens achieves held-out PPL 8.88-8.93 at >89% per-element sparsity, trailing GPT-2 201M by 7.7% while showing that LIF temporal integration outperforms simple top-k masking at matched sparsity.
PMNet uses unitary phasor dynamics and hierarchical anchors to make explicit memory stable for long sequences, matching a 3x larger Mamba model on long-context robustness with a 119M parameter network.
Transformer components arise as the natural solution to precision-weighted directional state estimation on the hypersphere.
HubRouter is a sub-quadratic routing primitive using learned hubs that replaces attention layers in hybrid models while delivering competitive perplexity and large throughput gains.
Recurrent Transformers add per-layer recurrent memory via self-attention on own activations plus a tiling algorithm that reduces training memory traffic, yielding better C4 pretraining cross-entropy than parameter-matched standard transformers with fewer layers.
Co-locating tests with implementation code yields substantially higher preservation and correctness in foundation-model-generated programs than separated test syntax.
The Linear Accessibility Profile predicts steering vector effectiveness and optimal layers with Spearman correlations of 0.86-0.91 using unembedding projections on intermediate states across multiple models and concepts.
Scal3R achieves better accuracy and consistency in large-scale 3D scene reconstruction by maintaining a compressed global context through test-time adaptation of lightweight neural networks on long video sequences.
Attention Editing converts pre-trained LLMs to new attention architectures through layer-wise teacher-forced optimization and model-level distillation, preserving performance with efficiency gains.
PAM, a complex-valued associative memory model, exhibits steeper power-law scaling in loss and perplexity than a matched real-valued baseline when trained on WikiText-103 from 5M to 100M parameters.
A two-stage distillation recipe converts a Pythia-1B Transformer into a Mamba model that preserves performance with perplexity 14.11 versus the teacher's 13.86.
citing papers explorer
-
When Does Content-Based Routing Work? Representation Requirements for Selective Attention in Hybrid Sequence Models
Content-based routing succeeds only when models provide bidirectional context and perform pairwise comparisons, with bidirectional Mamba plus rank-1 projection reaching 99.7% precision at linear inference cost.
-
Rotation Equivariant Mamba for Vision Tasks
EQ-VMamba adds rotation-equivariant cross-scan and group Mamba blocks to enforce end-to-end rotation equivariance, yielding better rotation robustness, competitive accuracy, and roughly 50% fewer parameters than non-equivariant baselines across classification, segmentation, and super-resolution.
-
Learning to (Learn at Test Time): RNNs with Expressive Hidden States
TTT layers treat the hidden state as a trainable model updated at test time, allowing linear-complexity sequence models to scale perplexity reduction with context length unlike Mamba.
-
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Mamba is a linear-time sequence model using input-dependent selective SSMs that achieves SOTA results across modalities and matches twice-larger Transformers on language modeling with 5x higher inference throughput.
-
Parallel Scan Recurrent Neural Quantum States for Scalable Variational Monte Carlo
PSR-NQS makes recurrent neural quantum states scalable for variational Monte Carlo by using parallel scan recurrence, reaching accurate results on 52x52 two-dimensional lattices.
-
Affinity Is Not Enough: Recovering the Free Energy Principle in Mixture-of-Experts
Adding temporal memory via LIF, precision-weighted gating, and anticipatory prediction to MoE routers recovers effective expert selection at distribution transitions, with ablation confirming a super-additive beta-ant interaction.
-
Sparse Prefix Caching for Hybrid and Recurrent LLM Serving
Sparse prefix caching via dynamic programming for optimal checkpoint placement under overlap distributions improves the Pareto frontier for recurrent and hybrid LLM serving on shared-prefix data.
-
Winner-Take-All Spiking Transformer for Language Modeling
Winner-take-all spiking self-attention replaces softmax in spiking transformers to support language modeling on 16 datasets with spike-driven, energy-efficient architectures.
-
ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training
ZipMap achieves linear-time bidirectional 3D reconstruction by zipping image collections into a compact stateful representation via test-time training layers.
-
DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoning
DELTA partitions layers into full, delta, and sparse groups to select salient tokens via aggregated attention scores, matching full-attention accuracy on AIME and GPQA while cutting attended tokens up to 4.25x and achieving 1.54x speedup.
-
FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision
FlashAttention-3 achieves 1.5-2x speedup on H100 GPUs for attention, reaching 740 TFLOPs/s (75% utilization) in FP16 and near 1.2 PFLOPs/s in FP8 while cutting numerical error by 2.6x versus baseline FP8 attention.
-
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Transformers and SSMs are unified through structured state space duality, producing a 2-8X faster Mamba-2 model that remains competitive with Transformers.
-
Discontinuous Galerkin Neural Operator for Pathology Defocus Deblurring
DGNO parameterizes integral kernels with discontinuous Galerkin elements for heterogeneous defocus deblurring in pathology images and reports superior performance over prior methods.
-
SymbolicLight V1: Spike-Gated Dual-Path Language Modeling with High Activation Sparsity and Sub-Billion-Scale Pre-Training Evidence
A 194M-parameter spiking dual-path model trained on 3B Chinese-English tokens achieves held-out PPL 8.88-8.93 at >89% per-element sparsity, trailing GPT-2 201M by 7.7% while showing that LIF temporal integration outperforms simple top-k masking at matched sparsity.
-
Phasor Memory Networks: Stable Backpropagation Through Time for Scalable Explicit Memory
PMNet uses unitary phasor dynamics and hierarchical anchors to make explicit memory stable for long sequences, matching a 3x larger Mamba model on long-context robustness with a 119M parameter network.
-
RT-Transformer: The Transformer Block as a Spherical State Estimator
Transformer components arise as the natural solution to precision-weighted directional state estimation on the hypersphere.
-
HubRouter: A Pluggable Sub-Quadratic Routing Primitive for Hybrid Sequence Models
HubRouter is a sub-quadratic routing primitive using learned hubs that replaces attention layers in hybrid models while delivering competitive perplexity and large throughput gains.
-
The Recurrent Transformer: Greater Effective Depth and Efficient Decoding
Recurrent Transformers add per-layer recurrent memory via self-attention on own activations plus a tiling algorithm that reduces training memory traffic, yielding better C4 pretraining cross-entropy than parameter-matched standard transformers with fewer layers.
-
Co-Located Tests, Better AI Code: How Test Syntax Structure Affects Foundation Model Code Generation
Co-locating tests with implementation code yields substantially higher preservation and correctness in foundation-model-generated programs than separated test syntax.
-
Predicting Where Steering Vectors Succeed
The Linear Accessibility Profile predicts steering vector effectiveness and optimal layers with Spearman correlations of 0.86-0.91 using unembedding projections on intermediate states across multiple models and concepts.
-
Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction
Scal3R achieves better accuracy and consistency in large-scale 3D scene reconstruction by maintaining a compressed global context through test-time adaptation of lightweight neural networks on long video sequences.
-
Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion
Attention Editing converts pre-trained LLMs to new attention architectures through layer-wise teacher-forced optimization and model-level distillation, preserving performance with efficiency gains.
-
Phase-Associative Memory: Sequence Modeling in Complex Hilbert Space
PAM, a complex-valued associative memory model, exhibits steeper power-law scaling in loss and perplexity than a matched real-valued baseline when trained on WikiText-103 from 5M to 100M parameters.
-
Attention to Mamba: A Recipe for Cross-Architecture Distillation
A two-stage distillation recipe converts a Pythia-1B Transformer into a Mamba model that preserves performance with perplexity 14.11 versus the teacher's 13.86.
-
M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling
M²RNN achieves perfect state tracking at unseen lengths and outperforms Gated DeltaNet hybrids by 0.4-0.5 perplexity on 7B models with 3x smaller recurrent states.
-
On the Spatiotemporal Dynamics of Generalization in Neural Networks
Deriving a neural cellular automaton from locality, symmetry, and stability postulates produces 100% accurate addition generalization from 16-digit to 1-million-digit inputs.
-
LinMU: Multimodal Understanding Made Linear
LinMU achieves linear-complexity multimodal understanding by swapping self-attention for an M-MATE dual-branch block and distilling from a frozen teacher VLM, matching accuracy with up to 2.7x faster TTFT and 9x higher throughput.
-
Lizard: An Efficient Linearization Framework for Large Language Models
Lizard linearizes Transformer LLMs via subquadratic attention and adaptive learnable modules, recovering near-original performance while outperforming prior linearization methods on MMLU and associative recall.
-
MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent
MemAgent uses multi-conversation RL to train a memory agent that reads text in segments and overwrites memory, extrapolating from 8K training to 3.5M token QA with under 5% loss and 95%+ on 512K RULER.
-
Mercury: Ultra-Fast Language Models Based on Diffusion
Mercury Coder diffusion LLMs achieve throughputs of 1109 and 737 tokens per second on H100 GPUs, up to 10x faster than frontier models with comparable quality.
-
MoBA: Mixture of Block Attention for Long-Context LLMs
MoBA routes attention over blocks via MoE-style gating to enable dynamic, bias-light long-context attention that matches full attention performance at lower cost.
-
LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation
LightTransfer identifies lazy layers in LLMs like LLaMA and replaces their attention with streaming attention to form hybrid models, delivering up to 2.17x throughput with under 1.5% drop on LongBench and strong results on reasoning benchmarks.
-
Gated Linear Attention Transformers with Hardware-Efficient Training
Gated linear attention Transformers achieve competitive language modeling results with linear-time inference, superior length generalization, and higher training throughput than Mamba.
-
The Falcon Series of Open Language Models
Falcon-180B is a 180B-parameter open decoder-only model trained on 3.5 trillion tokens that approaches PaLM-2-Large performance at lower cost and is released with dataset extracts.
-
Parallel Recursive LSTM
PR-LSTM replaces linear recurrence with recursive gated merging over a balanced binary tree to achieve log-depth parallelism without restricting transitions to linear or associative forms.
-
Kaczmarz Linear Attention
Kaczmarz Linear Attention replaces the empirical coefficient in Gated DeltaNet with a key-norm-normalized step size derived from the online regression objective, yielding lower perplexity and better needle-in-haystack performance.
-
MDN: Parallelizing Stepwise Momentum for Delta Linear Attention
MDN parallelizes stepwise momentum for delta linear attention using geometric reordering and dynamical systems analysis, yielding performance gains over Mamba2 and GDN on 400M and 1.3B models.
-
Absorber LLM: Harnessing Causal Synchronization for Test-Time Training
Absorber LLM introduces causal synchronization to absorb context into parameters for memory-efficient long-context LLM inference while preserving causal effects.
-
The Cognitive Penalty: Ablating System 1 and System 2 Reasoning in Edge-Native SLMs for Decentralized Consensus
System 1 intuition in edge SLMs delivers 100% adversarial robustness and low latency for DAO consensus while System 2 reasoning causes 26.7% cognitive collapse and 17x slowdown.
-
Adaptive Spiking Neurons for Vision and Language Modeling
ASN uses trainable parameters for adaptive membrane dynamics and firing in SNNs, with NASN adding normalization, and reports effectiveness across 19 vision and language datasets.
-
Belief-State RWKV for Reinforcement Learning under Partial Observability
Belief-state RWKV maintains an uncertainty-aware recurrent state for RL policies in partial observability and shows modest gains over standard recurrent baselines in a pilot with observation noise.
-
MFC-RFNet: A Multi-scale Guided Rectified Flow Network for Radar Sequence Prediction
MFC-RFNet integrates multi-scale bidirectional communication, condition-guided alignment, and rectified flow to produce clearer and more skillful radar precipitation forecasts than prior baselines on four public datasets.
-
Nirvana: A Specialized Generalist Model With Task-Aware Memory Mechanism
Nirvana adds a task-aware memory trigger and updater to specialized generalist models, achieving strong general benchmark results, lowest perplexity in biomedicine/finance/law, and improved MRI reconstruction fidelity.
-
StateX: Enhancing RNN Recall via Post-training State Expansion
StateX post-trains RNNs to expand recurrent state size, improving recall and in-context learning with negligible parameter growth.
-
Representation Paradigms in AI-based 3D Radiological Image Reconstruction: A Systematic Review
A systematic review that categorizes AI-based 3D radiological image reconstruction algorithms into four representation paradigms, summarizes evaluation metrics and datasets, and outlines challenges and future directions.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.
-
Large Language Models: A Survey
The paper surveys key large language models, their training methods, datasets, evaluation benchmarks, and future research directions in the field.
-
A Survey of Large Language Models
This survey reviews the background, key techniques, and evaluation methods for large language models, emphasizing emergent abilities that appear at large scales.
-
Evolution of Video Generative Foundations
This survey traces video generation technology from GANs to diffusion models and then to autoregressive and multimodal approaches while analyzing principles, strengths, and future trends.
-
Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices
Position paper claiming that distributed training across massive edge devices can overcome data depletion and centralized compute monopolies in LLM scaling.