Mesh inference allows a network of agents to reach the centralized optimum through local relaxations of a coupled free energy using only admitted observations, with convergence guaranteed by M-matrix properties in the linear-Gaussian regime.
hub Canonical reference
Ring Attention with Blockwise Transformers for Near-Infinite Context
Canonical reference. 94% of citing Pith papers cite this work as background.
59
Pith papers citing it
12
external citations · Pith
Background
94% of classified citations
abstract
Transformers have emerged as the architecture of choice for many state-of-the-art AI models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands imposed by Transformers limit their ability to handle long sequences, thereby posing challenges in utilizing videos, actions, and other long-form sequences and modalities in complex environments. We present a novel approach, Ring Attention with Blockwise Transformers (Ring Attention), which leverages blockwise computation of self-attention and feedforward to distribute long sequences across multiple devices while fully overlapping the communication of key-value blocks with the computation of blockwise attention. Our approach enables training and inference of sequences that are up to device count times longer than those achievable by prior memory-efficient Transformers, without resorting to approximations or incurring additional communication and computation overheads. Extensive experiments on language modeling and reinforcement learning tasks demonstrate the effectiveness of our approach in allowing millions of tokens context size and improving performance.
hub tools
citation-role summary
background 17
dataset 1
citation-polarity summary
representative citing papers
ODB is an online batching system for distributed LLM training that forms batches post-preprocessing, provides formal deadlock-free guarantees via the Distributed Group Alignment Problem, and reports 1.58-3.78x throughput gains versus fixed-batch baselines.
Tensor Cache augments sliding-window attention with an eviction-fed outer-product associative memory and a training correction to improve long-context performance under bounded memory.
LongLive-2.0 delivers an NVFP4 parallel infrastructure that enables direct training of long multi-shot autoregressive diffusion video models and achieves up to 2.15x training and 1.84x inference speedups on Blackwell and other GPUs.
Proves that RoPE attention loses locality bias and token distinction in long contexts, approaching random behavior independent of content.
Lighthouse Attention enables faster long-context pre-training via gradient-free symmetrical hierarchical compression of QKV while preserving causality, followed by a short full-attention recovery that yields lower loss than standard full-attention training.
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
MoE-Prefill achieves 1.35-1.59x higher throughput for prefill-only MoE serving by using asynchronous expert parallelism to overlap weight AllGather with computation and prefix-aware routing with true-FLOPs tracking.
PrefixWall mitigates APC side channels in multi-tenant LLM systems via selective prefix isolation, delivering up to 70% higher cache reuse and 30% lower latency than full-isolation baselines.
Exact Flow Linear Attention derives a closed-form exact update for delta-rule linear attention from continuous-time dynamics, removing Euler discretization error while preserving linear complexity and structure.
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
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.
Infini-attention combines compressive memory with masked local attention and long-term linear attention inside each Transformer block to support infinite context length with bounded resources.
Synthetic page-ranked reasoning traces plus low-strength model merging give a 32B VLM 58.3 on MMLongBenchDoc, beating a 235B teacher while cutting output tokens ~12× versus explicit reasoning.
PHOENIX recovers permanent node failures in LLM training via hot-swapping of spares using zero-overhead per-step in-memory optimizer-state replication, finishing recovery in under 40 s on up to 512 GPUs.
A CPU-GPU hybrid design with stream-loading prefill, expert parallelism, and disaggregation achieves cloud SLOs for local MoE inference on dual-socket CPUs and consumer GPUs.
FlashCP introduces Whole-Doc sharding, sharding-aware KV communication, and a heuristic for mixed sharding plans, claiming up to 1.63x speedup over prior CP methods for LLM training.
FedRAG uses a Scrambled Distributed Attention protocol with feature scrambling and token permutation to enable high-throughput, privacy-preserving federated RAG without special hardware or retraining.
A supervised fine-tuning approach using inverted multi-resolution planning scaffolds from public-domain novels trains models to generate book-length stories with more human-like literary qualities than standard instruction-tuned LLMs.
ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.
EXACT re-allocates training supervision by inverse frequency of long effective-context targets, improving NoLiMa and RULER scores by 5-18 points on Qwen and LLaMA models without degrading standard QA or reasoning.
FocuSFT uses an inner optimization loop to adapt fast-weight parameters into a parametric memory that sharpens attention on relevant content, then conditions outer-loop supervised fine-tuning on this representation, yielding gains on long-context benchmarks.
ZAYA1-8B is a reasoning MoE model with 700M active parameters that matches larger models on math and coding benchmarks and reaches 91.9% on AIME'25 via Markovian RSA test-time compute.
citing papers explorer
-
Mesh Inference: A Formal Model of Collective Inference Without a Center
Mesh inference allows a network of agents to reach the centralized optimum through local relaxations of a coupled free energy using only admitted observations, with convergence guaranteed by M-matrix properties in the linear-Gaussian regime.
-
Online Dynamic Batching with Formal Guarantees for LLM Training
ODB is an online batching system for distributed LLM training that forms batches post-preprocessing, provides formal deadlock-free guarantees via the Distributed Group Alignment Problem, and reports 1.58-3.78x throughput gains versus fixed-batch baselines.
-
Tensor Cache: Eviction-conditioned Associative Memory for Transformers
Tensor Cache augments sliding-window attention with an eviction-fed outer-product associative memory and a training correction to improve long-context performance under bounded memory.
-
LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation
LongLive-2.0 delivers an NVFP4 parallel infrastructure that enables direct training of long multi-shot autoregressive diffusion video models and achieves up to 2.15x training and 1.84x inference speedups on Blackwell and other GPUs.
-
RoPE Distinguishes Neither Positions Nor Tokens in Long Contexts, Provably
Proves that RoPE attention loses locality bias and token distinction in long contexts, approaching random behavior independent of content.
-
Long Context Pre-Training with Lighthouse Attention
Lighthouse Attention enables faster long-context pre-training via gradient-free symmetrical hierarchical compression of QKV while preserving causality, followed by a short full-attention recovery that yields lower loss than standard full-attention training.
-
SCOUT: Active Information Foraging for Long-Text Understanding with Decoupled Epistemic States
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
-
MoE-Prefill: Zero Redundancy Overheads in MoE Prefill Serving
MoE-Prefill achieves 1.35-1.59x higher throughput for prefill-only MoE serving by using asynchronous expert parallelism to overlap weight AllGather with computation and prefix-aware routing with true-FLOPs tracking.
-
PrefixWall: Mitigating Prefix Caching Side Channels in Shared LLM Systems
PrefixWall mitigates APC side channels in multi-tenant LLM systems via selective prefix isolation, delivering up to 70% higher cache reuse and 30% lower latency than full-isolation baselines.
-
Exact Flow Linear Attention: Exact Solution from Continuous-Time Dynamics
Exact Flow Linear Attention derives a closed-form exact update for delta-rule linear attention from continuous-time dynamics, removing Euler discretization error while preserving linear complexity and structure.
-
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
-
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.
-
Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
Infini-attention combines compressive memory with masked local attention and long-term linear attention inside each Transformer block to support infinite context length with bounded resources.
-
Internalized Reasoning for Long-Context Visual Document Understanding
Synthetic page-ranked reasoning traces plus low-strength model merging give a 32B VLM 58.3 on MMLongBenchDoc, beating a 235B teacher while cutting output tokens ~12× versus explicit reasoning.
-
PHOENIX: Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint
PHOENIX recovers permanent node failures in LLM training via hot-swapping of spares using zero-overhead per-step in-memory optimizer-state replication, finishing recovery in under 40 s on up to 512 GPUs.
-
Achieving Cloud-Grade SLOs for Local Mixture-of-Experts Inference through CPU-GPU Hybrid Design
A CPU-GPU hybrid design with stream-loading prefill, expert parallelism, and disaggregation achieves cloud SLOs for local MoE inference on dual-socket CPUs and consumer GPUs.
-
FlashCP: Load-Balanced Communication-Efficient Context Parallelism for LLM Training
FlashCP introduces Whole-Doc sharding, sharding-aware KV communication, and a heuristic for mixed sharding plans, claiming up to 1.63x speedup over prior CP methods for LLM training.
-
An Efficient and Privacy-Preserving Architecture for Cross-Institutional Collaborative RAG
FedRAG uses a Scrambled Distributed Attention protocol with feature scrambling and token permutation to enable high-throughput, privacy-preserving federated RAG without special hardware or retraining.
-
Towards Human-Level Book-Writing Capability
A supervised fine-tuning approach using inverted multi-resolution planning scaffolds from public-domain novels trains models to generate book-length stories with more human-like literary qualities than standard instruction-tuned LLMs.
-
ShardTensor: Domain Parallelism for Scientific Machine Learning
ShardTensor is a domain-parallelism system for SciML that enables flexible scaling of extreme-resolution spatial datasets by removing the constraint of batch size one per device.
-
Where Does Long-Context Supervision Actually Go? Effective-Context Exposure Balancing
EXACT re-allocates training supervision by inverse frequency of long effective-context targets, improving NoLiMa and RULER scores by 5-18 points on Qwen and LLaMA models without degrading standard QA or reasoning.
-
FocuSFT: Bilevel Optimization for Dilution-Aware Long-Context Fine-Tuning
FocuSFT uses an inner optimization loop to adapt fast-weight parameters into a parametric memory that sharpens attention on relevant content, then conditions outer-loop supervised fine-tuning on this representation, yielding gains on long-context benchmarks.
-
ZAYA1-8B Technical Report
ZAYA1-8B is a reasoning MoE model with 700M active parameters that matches larger models on math and coding benchmarks and reaches 91.9% on AIME'25 via Markovian RSA test-time compute.
-
The Impossibility Triangle of Long-Context Modeling
No model can achieve efficiency, compactness, and recall capacity scaling with sequence length at once, as any two imply a strict bound of O(poly(d)/log V) on recallable facts.
-
Unifying Sparse Attention with Hierarchical Memory for Scalable Long-Context LLM Serving
SPIN co-designs sparse attention with hierarchical memory to achieve 1.66-5.66x higher throughput, 7-9x lower TTFT, and up to 58% lower TPOT than vLLM and original sparse implementations.
-
ChipLight: Cross-Layer Optimization of Chiplet Design with Optical Interconnects for LLM Training
ChipLight is a multi-objective optimization framework that co-designs chiplet hardware, training parallelism, and optical networks to improve efficiency in distributed LLM training clusters.
-
Where to Focus: Query-Modulated Multimodal Keyframe Selection for Long Video Understanding
Q-Gate dynamically routes keyframe selection in long videos via query-modulated gating across visual grounding, global matching, and contextual alignment experts to improve MLLM performance.
-
Hive: A Multi-Agent Infrastructure for Algorithm- and Task-Level Scaling
Hive is a multi-agent infrastructure with a logits cache for reducing cross-path redundancy in sampling and agent-aware scheduling for better compute and KV-cache allocation, shown to deliver 1.11x-1.76x speedups and 33%-51% lower hotspot miss rates.
-
CoCoDiff: Optimizing Collective Communications for Distributed Diffusion Transformer Inference Under Ulysses Sequence Parallelism
CoCoDiff achieves 3.6x average and 8.4x peak speedup for distributed DiT inference on up to 96 GPU tiles via tile-aware all-to-all, V-first scheduling, and selective V communication.
-
LSRM: High-Fidelity Object-Centric Reconstruction via Scaled Context Windows
Scaling transformer context with sparse attention and 3D-aware block routing improves feed-forward 3D reconstruction and inverse rendering, closing much of the quality gap with dense-view optimization.
-
DeepStack: Scalable and Accurate Design Space Exploration for Distributed 3D-Stacked AI Accelerators
DeepStack introduces a fast performance model and hierarchical search method for co-optimizing 3D DRAM stacking, interconnects, and distributed scheduling in AI accelerators, delivering up to 9.5x throughput gains over baselines.
-
GENSERVE: Efficient Co-Serving of Heterogeneous Diffusion Model Workloads
GENSERVE improves SLO attainment by up to 44% for co-serving heterogeneous T2I and T2V diffusion workloads via step-level preemption, elastic parallelism, and joint scheduling.
-
Kling-Omni Technical Report
Kling-Omni is a unified multimodal generative system that produces cinematic videos from diverse inputs by integrating generation, editing, and intelligent reasoning in a single end-to-end model.
-
InfiniPipe: Elastic Pipeline Parallelism for Efficient Variable-Length Long-Context LLM Training
InfiniPipe proposes elastic pipeline parallelism and stage-aware chunk-level adaptive checkpointing to achieve 1.69x speedup over state-of-the-art for variable-length long-context LLM training.
-
MAGI-1: Autoregressive Video Generation at Scale
MAGI-1 is a 24B-parameter autoregressive video world model that predicts denoised frame chunks sequentially with increasing noise to enable causal, scalable, streaming generation up to 4M token contexts.
-
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.
-
PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference
PipeFusion applies patch partitioning and pipeline parallelism with one-step stale feature reuse to reduce communication overhead in DiT inference, reporting SOTA results on 8x L40 GPUs for Pixart, SD3, and Flux.1.
-
OpenRLHF: An Easy-to-use, Scalable and High-performance RLHF Framework
OpenRLHF is a new open-source RLHF framework reporting 1.22x to 1.68x speedups and fewer lines of code than prior systems.
-
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.
-
Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE
Jet-Long is a tuning-free bifocal RoPE method that dynamically sets remote group size from sequence length, recovering the base model within the pretrained window and beating prior zero-shot extenders on RULER, HELMET-RAG, and PG-19 with near-FA2 throughput.
-
KernelFlume: Elastic Core-Attention Scaling for Agentic Long-Context Decoding
KernelFlume presents a disaggregated decode architecture that separates core attention from projection/FFN paths to enable elastic scaling of attention nodes, reporting up to 61% lower cost per million tokens versus full-instance scaling on H100 hardware for Llama-3.1-8B under dynamic long-context w
-
torchtune: PyTorch native post-training library
torchtune is a modular PyTorch library for LLM post-training that delivers competitive performance and memory efficiency while supporting rapid research iteration through hackable components.
-
Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents
A dual-process memory architecture for scientific AI agents maintains 70-85% accuracy over 15,000 messages by using a constant 10-message episodic window and domain-specific semantic consolidation, consuming 62% fewer tokens than full-context baselines.
-
MemReread: Enhancing Agentic Long-Context Reasoning via Memory-Guided Rereading
MemReread improves agent long-context reasoning by triggering rereading on insufficient final memory to recover discarded indirect facts, outperforming baselines at linear complexity.
-
Unleashing Scalable Context Parallelism for Foundation Models Pre-Training via FCP
FCP shards sequences at block level with flexible P2P communication and bin-packing to achieve near-linear scaling up to 256 GPUs and 1.13x-2.21x higher attention MFU in foundation model pre-training.
-
An Efficient Hybrid Sparse Attention with CPU-GPU Parallelism for Long-Context Inference
Fluxion achieves 1.5x-3.7x speedup in long-context LLM inference with CPU KV caches while limiting accuracy degradation to at most 0.26 relative to full attention.
-
StreamIndex: Memory-Bounded Compressed Sparse Attention via Streaming Top-k
Chunked streaming top-k enables CSA indexer execution at 1M sequence length with 6.21 GB peak memory and >=0.998 recall on synthetic V4-shaped inputs.
-
InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling
InternVideo2.5 improves video MLLMs by incorporating dense vision task annotations via direct preference optimization and compact spatiotemporal representations via adaptive hierarchical token compression, yielding better benchmark performance, 6x longer video memory, and new capabilities likeobject
-
HunyuanVideo: A Systematic Framework For Large Video Generative Models
HunyuanVideo presents a 13B-parameter open-source video generative model with integrated data, architecture, training, and inference systems whose professional evaluations show it outperforming prior SOTA models including Runway Gen-3 and Luma 1.6.