Pith. sign in

Paper Citation Record · LEDGER

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

As of 10 August 2026, this Paper Citation Record lists 14 of 14 outbound references and 9 inbound Pith citation observations for arXiv:2502.01563.

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

pith.paper-citation-record.v1
2502.01563 v4

Coverage vector

measured 14 of 14 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-09T15:03:02.760143Z

measured 23 of 23 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-09T06:31:02.800959+00:00

measured 9 of 9 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-12T05:46:45.293902Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-04T09:09:43.221590Z

Reference resolution

14 of 14 outbound references displayed

  • verified exact0
  • verified fuzzy2
  • unresolved8
  • parse uncertain0
  • malformed identifier4
  • metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation eac6b215-9d94-4e80-a9bc-35d47158a67a · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 1

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.896354Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.740813Z digest=sha256:b1e5e283d0747a7fb3f4e94f2c3db9a5a62a77bd64280256d96bdc5162482ac3

Observation 7bb69b89-ef7d-4537-8a5a-32cefabafac8 · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 2

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.954804Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.723148Z digest=sha256:786eb2d49acb515e1ce741a524c713b14e9a14c257dab329d44f1dc7a38e438c

Observation 0fb360ef-7249-4793-a7b9-3da879a3bb47 · outbound

This paper cites The Llama 3 Herd of Models.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding The Llama 3 Herd of Models

Reference 3

Resolution
unresolved
no resolver link, observed 2026-08-09T15:03:02.707504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-09T15:03:02.707504Z digest=sha256:9a4e2218dc040921797518fc3cffef5c7ff539e8e0818c2ad6fee838245fefc0

Observation 6148e671-a4fa-4de3-aba6-f57b586e165b · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 7

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.940726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.727583Z digest=sha256:3713251c8c25e0e4045b9e2f887acc070125b6cea024969737e3c939cdde0ca9

Observation 2b1c81e0-764b-4622-aa2f-b1ae3733338e · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 8

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.926564Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.732065Z digest=sha256:3aaa1c6c896bcd3ef796d7b48a4e0e85c10ab3bc44bd5618b70d057bdd15e5a5

Observation f464e44e-6a4d-47a4-8d01-1c4d321d11ff · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 9

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.911785Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.736466Z digest=sha256:acf58637446f5a3525c854956f8422b74d89ea185ff707dbc3da2cc281bde300

Observation 304ec517-ccc2-44f7-af01-58306916c156 · outbound

This paper cites add norm,.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding add norm,

Reference 11

Resolution
malformed identifier
raw_fallback, observed 2026-08-09T15:03:02.882752Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.745108Z digest=sha256:fd113c4f86d036184c989b8182ff7f0336cfd21d186187ca436470d7cb20f18f

Observation 8c4fcd3d-c30e-47ba-ac70-be1569100563 · outbound

This paper cites , wi−1) is extracted from the predicted probability tensor P by selecting the value corresponding to the actual token wi’s index in the vocabulary.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding , wi−1) is extracted from the predicted probability tensor P by selecting the value corresponding to the actual token wi’s index in the vocabulary

Reference 12

Resolution
verified fuzzy
raw_fallback, observed 2026-08-09T15:03:02.868906Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.750659Z digest=sha256:71e953810b79128d3f24e6544f5332728040611abd845adb23793edf89b38fa2

Observation 50a53ad2-ccc9-49f2-9601-055d8970a3de · outbound

This paper cites an unresolved cited work.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Unresolved cited work

Reference 13

Resolution
unresolved
raw_fallback, observed 2026-08-09T15:03:02.854186Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.755829Z digest=sha256:e629dc28f880e3c71b6c6324d7617a8f6bc9fa8ee4196d91491b41eb1d290816

Observation b2311d4d-c403-44fa-a33b-6dbc5aa78f44 · outbound

This paper cites poodlesque.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding poodlesque

Reference 14

Resolution
malformed identifier
raw_fallback, observed 2026-08-09T15:03:02.840344Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.760143Z digest=sha256:2788fbf10df0c31e7cf2d54e24f9ef4139ea5954c6b2530c4a9ba96aff501463

Observation 4f5670f6-5c38-4ead-a8c2-47688b547b50 · outbound

This paper cites Liu, H., Li, C., Wu, Q., and Lee, Y.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Liu, H., Li, C., Wu, Q., and Lee, Y

Reference 2017

Resolution
malformed identifier
raw_fallback, observed 2026-08-09T15:03:02.969634Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.712687Z digest=sha256:5c423e0950d37568af513797513095f103a0669d9123c33f86bbaaa9e7b28bf8

Observation 6a8d3667-ff6a-452c-937b-eba6b092d4b7 · outbound

This paper cites Dao, T., Fu, D.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Dao, T., Fu, D

Reference 2019

Resolution
verified fuzzy
raw_fallback, observed 2026-08-09T15:03:02.984904Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-08-09T15:03:02.702913Z digest=sha256:3fa47be6d00e73f6900efd307c1ea1ffb3603821a95573a5a2b966358ef19b17

Observation 94f3f7d4-0db5-4f15-a0bd-5206b888f86a · outbound

This paper cites Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models

Reference 2022

Resolution
unresolved
no resolver link, observed 2026-08-09T15:03:02.697258Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-09T15:03:02.697258Z digest=sha256:3bf7a92cbc3dff2a3306d7efb2af87b49d2268e1a0e3d942a2c5ef04c4dc2075

Observation ce1ad9c2-9d64-4a6c-8431-96dbc3d207ff · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding Gemma: Open Models Based on Gemini Research and Technology

Reference 2023

Resolution
malformed identifier
no resolver link, observed 2026-08-09T15:03:02.717734Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-09T15:03:02.717734Z digest=sha256:dac1e019fdb0909453ac7c64369abe571a09d5a2e46248fb4ca0e5802b0b3365

Pith citing papers

Observation bf93ba74-5279-4adf-8c14-ab37143ef9c2 · inbound

Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents cites this paper.

Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 109

Resolution
verified exact
arxiv_id, observed 2026-05-12T13:36:57.139599Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-05-12T13:36:57.011451Z digest=sha256:997fa0e349a41b5becbad074b6e468fee84b92e80937c45cab243e9ab29bbde9

Observation be338f70-addb-4805-a9ef-1db84a9d396a · inbound

LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning cites this paper.

LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 15

Resolution
verified exact
arxiv_id, observed 2026-05-10T11:20:10.533330Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=arxiv_source observed=2026-05-10T11:17:43.769244Z digest=sha256:85a3de8bdea352f7912163feb0403b40d252f45bb837adfe9b10577a8391a7d9

Observation 985854c6-e763-41a6-aca6-41b5241cf0cd · inbound

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

A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 9

Resolution
verified exact
arxiv_id, observed 2026-05-12T07:36:41.527286Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-12T02:29:20.796512Z digest=sha256:331d24c1b4524458e41c26489e2d5b384cd461cd8ba8f801fc83908c93331fa6

Observation 4a5771cc-95f9-49f5-b5ee-e1da544e81d2 · inbound

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

A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 9

Resolution
verified exact
arxiv_id, observed 2026-05-14T21:19:29.180766Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-14T21:03:25.624300Z digest=sha256:1d29a6bdda2b82983fae9762fd964a164ecdfe28cd3f846faf3d32f76cbb96a8

Observation 4030a12c-9df5-4cc2-9289-85fa8af370de · inbound

LoKA: Low-precision Kernel Applications for Recommendation Models At Scale cites this paper.

LoKA: Low-precision Kernel Applications for Recommendation Models At Scale Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 41

Resolution
verified exact
arxiv_id, observed 2026-05-12T06:06:28.029189Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-12T04:33:41.411292Z digest=sha256:7338b11cf0f575bb11694e8f348b898417220bbceaf079ea320e7600bea40b65

Observation 4e60b4aa-c5f9-4ed3-92ab-75ab388f872a · inbound

LoKA: Low-precision Kernel Applications for Recommendation Models At Scale cites this paper.

LoKA: Low-precision Kernel Applications for Recommendation Models At Scale Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 41

Resolution
verified exact
arxiv_id, observed 2026-05-15T04:59:46.092023Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-15T04:55:01.973832Z digest=sha256:cbf0dd47751a1d7b4f987f4814ef3deed9d49a3177f5af82bea50d5fcc19e2ec

Observation 40471e3a-fd86-4eab-96cd-5acd8235482b · inbound

OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond cites this paper.

OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 23

Resolution
verified exact
arxiv_id, observed 2026-05-20T07:58:07.496041Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-05-20T07:57:51.032025Z digest=sha256:8016716bb7303dd30cc12b044865716a7cf72974dac2d015fdf996a45799533c

Observation 9407c463-6e47-4544-9aab-b4d6064dcfa1 · inbound

AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent cites this paper.

AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 62

Resolution
verified exact
arxiv_id, observed 2026-07-04T09:09:43.223797Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.

source=pdf_text observed=2026-06-26T10:27:05.451888Z digest=sha256:24be4c367c98d9fdd07e2733ae3bfb64075e5a8cff1921ea37a61ce550c459e5

Observation 30aa4c26-1565-4d9d-a8d0-1338842f5912 · inbound

Awakening Diffusion Transformers: Eliciting Stronger Generation and Understanding via Massive Activation Modulation cites this paper.

Awakening Diffusion Transformers: Eliciting Stronger Generation and Understanding via Massive Activation Modulation Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Reference 40

Resolution
unresolved
no resolver link, observed 2026-07-12T05:46:45.293902Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-07-12T05:46:45.293902Z digest=sha256:253b22cbf0930c24b9e1170c42216d3001b83ba85a36fe8950172bc11cfa9e89