Paper Citation Record · LEDGER

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid

As of 8 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 2 inbound Pith citation observations for arXiv:2502.07563.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2502.07563 v1

Coverage vector

measured 37 of 37 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-08T12:25:42.790098Z

measured 39 of 39 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-01T17:32:37.570769Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-05-18T18:51:45.610699Z

Reference resolution

37 of 37 outbound references displayed

verified exact1
verified fuzzy5
unresolved30
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation b2761e0e-4b64-409e-af90-d7fe4de38bd5 · outbound

This paper cites GPT-4 Technical Report.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid GPT-4 Technical Report

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.612941Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1565af2-adb7-4c74-b980-67bf83a13cfe · outbound

This paper cites Linear Transformers with Learnable Kernel Functions are Better In-Context Models.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Linear Transformers with Learnable Kernel Functions are Better In-Context Models

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.624361Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3be2ee39-3959-4271-bd90-f009297dddef · outbound

This paper cites Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM's Translation Capability.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM's Translation Capability

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-08-08T12:25:43.257935Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 3ab4cdec-02e3-43ca-8111-9d3670376504 · outbound

This paper cites FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.640835Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91312a3a-7ed3-4b88-866a-fea054710de2 · outbound

This paper cites Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.646406Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c925f4d6-1bc0-4283-880d-3c95e3e0f5c8 · outbound

This paper cites The Llama 3 Herd of Models.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid The Llama 3 Herd of Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.656382Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7c14e75-2bbc-48cd-878f-db7fdcfb4804 · outbound

This paper cites Mamba: Linear-Time Sequence Modeling with Selective State Spaces.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Mamba: Linear-Time Sequence Modeling with Selective State Spaces

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.661139Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 834dc8f1-c38c-4d0e-a4bc-f1dc493a1ecc · outbound

This paper cites Measuring Massive Multitask Language Understanding.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Measuring Massive Multitask Language Understanding

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.666217Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc935757-5111-4368-955a-85e43c483e88 · outbound

This paper cites Repeat After Me: Transformers are Better than State Space Models at Copying.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Repeat After Me: Transformers are Better than State Space Models at Copying

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.671185Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ffb14d9-494f-4d45-a10e-7eb99a1d6501 · outbound

This paper cites MiniMax-01: Scaling Foundation Models with Lightning Attention.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid MiniMax-01: Scaling Foundation Models with Lightning Attention

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.680742Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b69a9752-719a-47e6-bd6d-a2636c0c7e99 · outbound

This paper cites Jamba: A Hybrid Transformer-Mamba Language Model.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Jamba: A Hybrid Transformer-Mamba Language Model

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.685122Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf67f9af-7e53-47d2-bdae-d883706bb468 · outbound

This paper cites an unresolved cited work.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-08T12:25:43.422421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 12a6d39f-d016-4439-87ba-0d22cc0d3524 · outbound

This paper cites doi: 10.18653/v1/2023.findings-emnlp.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid doi: 10.18653/v1/2023.findings-emnlp

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.693907Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c2fab43-7087-4322-8cde-93c3f06e6345 · outbound

This paper cites Dataset Decomposition: Faster LLM Training with Variable Sequence Length Curriculum.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Dataset Decomposition: Faster LLM Training with Variable Sequence Length Curriculum

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.703293Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b8e93ba-727c-40d1-abd2-01c8cd362928 · outbound

This paper cites TransNormerLLM: A Faster and Better Large Language Model with Improved TransNormer.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid TransNormerLLM: A Faster and Better Large Language Model with Improved TransNormer

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.707448Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 339ae241-9a41-4f7a-9c8e-ecafd7c78a05 · outbound

This paper cites Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.711865Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ee0d4a49-1b80-4720-9669-95881963ddff · outbound

This paper cites Scaling Laws for Linear Complexity Language Models.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Scaling Laws for Linear Complexity Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.721108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88ccc2bf-1a61-4aed-8e0d-f1c7062771e4 · outbound

This paper cites Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.725885Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7ec65b29-6216-4b59-9f1b-0d3a79743ac3 · outbound

This paper cites Linear Attention Sequence Parallelism.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Linear Attention Sequence Parallelism

Reference 25

Resolution: metadata mismatch; local_arxiv, observed 2026-08-08T12:25:42.952657Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation ab372280-d145-4bd1-91db-11abde4aaaf2 · outbound

This paper cites LongVILA: Scaling Long-Context Visual Language Models for Long Videos.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid LongVILA: Scaling Long-Context Visual Language Models for Long Videos

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.741468Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0ec0ec4-683c-44c3-ae7b-462d4e62bebd · outbound

This paper cites Gated Linear Attention Transformers with Hardware-Efficient Training.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Gated Linear Attention Transformers with Hardware-Efficient Training

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.746310Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2b0d8cd-b9c5-450a-8bba-bde9b09a046c · outbound

This paper cites Parallelizing Linear Transformers with the Delta Rule over Sequence Length.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Parallelizing Linear Transformers with the Delta Rule over Sequence Length

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.751276Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7a65e53-4525-4a5c-b54c-566a1621e8df · outbound

This paper cites Boosting Distributed Training Performance of the Unpadded BERT Model.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Boosting Distributed Training Performance of the Unpadded BERT Model

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.756424Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6760dd63-efe1-42db-ba9a-47bb5f14016b · outbound

This paper cites ByteTransformer: A high- performance transformer boosted for variable-length in- puts.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid ByteTransformer: A high- performance transformer boosted for variable-length in- puts

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T12:25:43.406108Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 63a8ebcd-779d-4711-b25d-16f3495823ca · outbound

This paper cites PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.771337Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 69562057-6f52-486b-a7b7-5ec38ad4514f · outbound

This paper cites As these techniques are variants of data parallelism, they integrate seamlessly with LASP.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid As these techniques are variants of data parallelism, they integrate seamlessly with LASP

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T12:25:43.389640Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 1571c3da-451f-4f1c-9c49-dcd8957877ae · outbound

This paper cites LASP-2 can manage variable sequence lengths efficiently by treating the entire batch as a single long sequence, streamlining the process without requiring padding.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid LASP-2 can manage variable sequence lengths efficiently by treating the entire batch as a single long sequence, streamlining the process without requiring padding

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T12:25:43.373969Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 3e494f39-e4b1-477b-881e-934873950d53 · outbound

This paper cites Split Size of Gathering 2048 512 128 32 Number of Splits 1 4 16 64 Throughput 486183 486166 486169 486158 A.5.4.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Split Size of Gathering 2048 512 128 32 Number of Splits 1 4 16 64 Throughput 486183 486166 486169 486158 A.5.4

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T12:25:43.340423Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation d3b1d05e-81bf-411b-bf0b-c8b773864e59 · outbound

This paper cites Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence

Reference 936

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.698693Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c48c134d-9f26-4c62-86c9-29f0f81fa76a · outbound

This paper cites An Empirical Study of Mamba-based Language Models.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid An Empirical Study of Mamba-based Language Models

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.735960Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32bc18e4-d811-4504-82ba-4ea714b00317 · outbound

This paper cites Gated Slot Attention for Efficient Linear-Time Sequence Modeling.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Gated Slot Attention for Efficient Linear-Time Sequence Modeling

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.766684Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a2c982bb-f607-469c-9c02-db98ad7db2a2 · outbound

This paper cites Adam: A Method for Stochastic Optimization.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Adam: A Method for Stochastic Optimization

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.676350Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98267f49-1450-4e61-b3ab-b1c491650e7d · outbound

This paper cites FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.716501Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 685225d0-e7d6-4c7b-a73b-8e6e7070c9d1 · outbound

This paper cites Fewer Truncations Improve Language Modeling.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Fewer Truncations Improve Language Modeling

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.651269Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6aee1e7-a230-4000-80da-79992ce586cb · outbound

This paper cites GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.618676Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 43ea1b62-b1b9-43a9-8ba9-e0a1cbe80823 · outbound

This paper cites Simple linear attention language models balance the recall-throughput tradeoff.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid Simple linear attention language models balance the recall-throughput tradeoff

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-08T12:25:42.629773Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca173ad3-351c-43fc-9841-3ce85014a41d · outbound

This paper cites L" denotes linear Transformer layers and.

LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid L" denotes linear Transformer layers and

Reference 2048

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T12:25:43.357187Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Pith citing papers

Observation 78e836d7-64ee-4e55-ab47-b3727d46e310 · inbound

SpikingBrain: Spiking Brain-inspired Large Models cites this paper.

SpikingBrain: Spiking Brain-inspired Large Models LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid

Reference 32

Resolution: verified exact; arxiv_id, observed 2026-05-18T18:51:45.613221Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 2967adc6-a89e-4b02-aa21-be8e8d1c471d · inbound

A Training-Memory Regression in MLA Sequence Parallelism: Why Megatron-Core Forbids Absorption, and LAGA -- a Communication-Efficient Fix cites this paper.

A Training-Memory Regression in MLA Sequence Parallelism: Why Megatron-Core Forbids Absorption, and LAGA -- a Communication-Efficient Fix LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T17:32:37.570769Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.