Paper Citation Record · LEDGER

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

As of 19 August 2026, this Paper Citation Record lists 35 of 35 outbound references and 0 inbound Pith citation observations for arXiv:2608.12149.

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

pith.paper-citation-record.v1
2608.12149 v1

Coverage vector

measured 35 of 35 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T00:18:35.310454Z

measured 35 of 35 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

35 of 35 outbound references displayed

verified exact1
verified fuzzy11
unresolved23
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation a3e4f8ee-7b0c-442d-a198-3843e0ce60c1 · outbound

This paper cites Systematic Outliers in Large Language Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Systematic Outliers in Large Language Models

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.644968Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3112966e-095e-4839-aad6-95f7a8001afb · outbound

This paper cites Qwen3-Coder-Next Technical Report.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Qwen3-Coder-Next Technical Report

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.722302Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5ebc945-e599-412a-8cab-0aaef105baa3 · outbound

This paper cites an unresolved cited work.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Unresolved cited work

Reference 5

Resolution: unresolved; raw_fallback, observed 2026-08-16T00:18:35.865952Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 81e1fd05-f38e-46e3-8916-68448e4f8859 · outbound

This paper cites No Language Left Behind: Scaling Human-Centered Machine Translation.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus No Language Left Behind: Scaling Human-Centered Machine Translation

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.740775Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7d39d2e4-ab2e-4fa0-807e-4b3fc554fa30 · outbound

This paper cites The Zamba2 Suite: Technical Report.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus The Zamba2 Suite: Technical Report

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.748692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d5a2055-c360-44c3-9db5-5998aee115e3 · outbound

This paper cites Summer is warm. Winter is cold.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Summer is warm. Winter is cold

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:35.760141Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 7ee92d5a-5642-4717-b0ee-d0a87341525a · outbound

This paper cites CodeSearchNet Challenge: Evaluating the State of Semantic Code Search.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus CodeSearchNet Challenge: Evaluating the State of Semantic Code Search

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.887327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7bb65d9d-9b7b-4688-a475-873bb44e2299 · outbound

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

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Jamba: A Hybrid Transformer-Mamba Language Model

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.897393Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a302ee69-6b44-45a7-adb1-7f83887a5629 · outbound

This paper cites Openceres: When open information extrac- tion meets the semi-structured web.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Openceres: When open information extrac- tion meets the semi-structured web

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.388985Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation fafa363d-1695-48a8-bb41-1f8f3d0a9ef1 · outbound

This paper cites Pointer Sentinel Mixture Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Pointer Sentinel Mixture Models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.905921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9db1afc5-9de9-47e2-a091-0ae47a10a8e1 · outbound

This paper cites A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-08-16T00:18:35.528311Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0995e174-6174-4423-8e46-36b4fc2fec4d · outbound

This paper cites Kvsink: Understanding and enhancing the preservation of attention sinks in kv cache quantization for llms.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Kvsink: Understanding and enhancing the preservation of attention sinks in kv cache quantization for llms

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.264822Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5d2fe05f-6c5f-44c8-a6b7-39cba0231592 · outbound

This paper cites Massive Activations in Large Language Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Massive Activations in Large Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.048113Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation acc87cef-edfd-409a-922c-52d187060696 · outbound

This paper cites The spike, the sparse and the sink: Anatomy of massive activations and attention sinks.arXiv preprint arXiv:2603.05498,.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus The spike, the sparse and the sink: Anatomy of massive activations and attention sinks.arXiv preprint arXiv:2603.05498,

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.070942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 14dddc32-3e14-45a6-8519-315543d69a32 · outbound

This paper cites Retentive Network: A Successor to Transformer for Large Language Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Retentive Network: A Successor to Transformer for Large Language Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.075087Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f92748c3-ef65-44b7-b624-05681a460ac4 · outbound

This paper cites Kimi Linear: An Expressive, Efficient Attention Architecture.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Kimi Linear: An Expressive, Efficient Attention Architecture

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.080028Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 63c1c30f-70c4-4998-bb6c-5a09dc94086e · outbound

This paper cites Efficient streaming language models with attention sinks.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Efficient streaming language models with attention sinks

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.197953Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 2c8b2b09-17d6-415c-a2f3-ee992ed281fc · outbound

This paper cites Exploring layer-wise information effectiveness for post- training quantization in small language models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Exploring layer-wise information effectiveness for post- training quantization in small language models

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.183353Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 84c599d5-2be7-49f9-a228-95e0a16947ef · outbound

This paper cites Qwen3 Technical Report.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Qwen3 Technical Report

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.100093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ffbc3440-38ca-4ee5-981b-2cf3dcb09cf8 · outbound

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

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Gated Linear Attention Transformers with Hardware-Efficient Training

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.103875Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b90a2a02-babf-40a6-bf41-50c4cc34ca8c · outbound

This paper cites Gated delta networks: Improving mamba2 with delta rule.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Gated delta networks: Improving mamba2 with delta rule

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.168168Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 27f9d213-eb58-45fd-8c30-31fe6e1d74e4 · outbound

This paper cites Beyond outliers: A data-free layer-wise mixed- precision quantization approach driven by numerical and structural dual-sensitivity.Under review, 2026a.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Beyond outliers: A data-free layer-wise mixed- precision quantization approach driven by numerical and structural dual-sensitivity.Under review, 2026a

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.053575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation e1b6f645-94c0-43a0-9a7d-a478ae67af47 · outbound

This paper cites Summer is warm. Winter is cold.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Summer is warm. Winter is cold

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:35.939669Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation be2d2464-8e88-4fd1-aa93-5d11a0f77b69 · outbound

This paper cites Pronounced MAs concentrate at attention-sink positions, particu- larly the initial token, “Summer,” and the first period.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Pronounced MAs concentrate at attention-sink positions, particu- larly the initial token, “Summer,” and the first period

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:35.924118Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 2cfc315c-9fcc-4277-802e-3c3332daa3c4 · outbound

This paper cites Summer is warm. Winter is cold.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Summer is warm. Winter is cold

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:35.773289Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f27c287a-5beb-4819-a9db-1393a63cfd01 · outbound

This paper cites A Systematic Analysis of Hybrid Linear Attention.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus A Systematic Analysis of Hybrid Linear Attention

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:35.084650Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18b9ebf3-aa6b-4d70-8e6d-9ca211bb544f · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Training Verifiers to Solve Math Word Problems

Reference 2018

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.731704Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 857cc832-55c3-4e70-9512-ac5beb451c75 · outbound

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

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus MiniMax-01: Scaling Foundation Models with Lightning Attention

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.892512Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 10db5d10-1117-4b87-81fd-f27751b72f8d · outbound

This paper cites Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.717169Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 18342080-f70e-42a4-8dfd-c51c76f33186 · outbound

This paper cites A discourse-aware attention model for abstractive summarization of long documents.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus A discourse-aware attention model for abstractive summarization of long documents

Reference 2021

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T00:18:36.401741Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 95ea3654-2dba-4d5b-bcf2-7d3ea87be02f · outbound

This paper cites Hidden dynamics of massive activations in transformer training.arXiv preprint arXiv:2508.03616,.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Hidden dynamics of massive activations in transformer training.arXiv preprint arXiv:2508.03616,

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.744844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation de5391b3-9799-4432-a250-c305d3960c86 · outbound

This paper cites RULER: What's the Real Context Size of Your Long-Context Language Models?.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus RULER: What's the Real Context Size of Your Long-Context Language Models?

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.882892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7a8fe65-24aa-4ba7-b896-2672290f103c · outbound

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

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Mamba: Linear-Time Sequence Modeling with Selective State Spaces

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.790267Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cde16a2e-2a9b-48ae-b816-d593fcaf3994 · outbound

This paper cites Just read twice: closing the recall gap for recurrent language models.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Just read twice: closing the recall gap for recurrent language models

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.712201Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32abbfe9-7ab3-4e07-bba0-b54781aa3997 · outbound

This paper cites Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge.

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-16T00:18:34.726669Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.