Paper Citation Record · LEDGER

Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

As of 4 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 13 inbound Pith citation observations for arXiv:2504.20966.

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

pith.paper-citation-record.v1
2504.20966 v4

Coverage vector

measured 0 of 0 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links

measured 13 of 13 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+00:00

measured 13 of 13 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-03T05:50:25.905992Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-06-29T08:23:15.622080Z

Reference resolution

0 of 0 outbound references displayed

verified exact0
verified fuzzy0
unresolved0
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

No outbound reference observations are available for this paper version.

Pith citing papers

Observation 51a088e9-3839-42ae-b99a-df800b24805a · inbound

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free cites this paper.

Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 36

Resolution: metadata mismatch; local_arxiv, observed 2026-05-12T09:04:34.829561Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e5eecf65-ba09-4cc8-b435-1164e311c6c7 · inbound

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse cites this paper.

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-05-16T08:47:37.236481Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 76878d18-181b-4c31-9292-092af760ca6a · inbound

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse cites this paper.

Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-03T05:50:25.905992Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc43e462-537f-4ce7-9475-27487d18bbfa · inbound

Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers cites this paper.

Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 50

Resolution: malformed identifier; local_arxiv, observed 2026-05-15T09:35:22.453752Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 1639b654-9aab-48e4-bd4f-b3999b366418 · inbound

Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation cites this paper.

Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 77

Resolution: verified exact; local_arxiv, observed 2026-05-11T09:05:58.105206Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 0b940071-93a4-4772-8f29-65fc64ccfe17 · inbound

FLUID: Continuous-Time Hyperconnected Sparse Transformer for Sink-Free Learning cites this paper.

FLUID: Continuous-Time Hyperconnected Sparse Transformer for Sink-Free Learning Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-05-11T17:21:09.954492Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6c0f6fdb-5e5e-4dca-afc5-08a1b296bf19 · 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 Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-05-12T07:36:41.426279Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 07a12766-33dd-480c-9d6c-4a69d7802684 · 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 Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 30

Resolution: verified exact; local_arxiv, observed 2026-05-14T21:19:29.078717Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d7f6adae-a93d-4835-9983-de9431a3a23a · inbound

Attention Sinks and Outliers in Attention Residuals cites this paper.

Attention Sinks and Outliers in Attention Residuals Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-20T12:13:16.040745Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6cb40198-ed7b-470f-adce-4f181def8767 · 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 Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 79

Resolution: verified exact; local_arxiv, observed 2026-05-20T07:58:07.603107Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation bda6fcf9-4e72-45fa-aac9-e892414412e3 · inbound

Most Transformer Modifications Still Do Not Transfer at 1-3B: A 2020-2026 Update to Narang et al. (2021) with Downstream Evaluation and a Noise Floor cites this paper.

Most Transformer Modifications Still Do Not Transfer at 1-3B: A 2020-2026 Update to Narang et al. (2021) with Downstream Evaluation and a Noise Floor Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 52

Resolution: verified exact; local_arxiv, observed 2026-05-21T06:19:42.075085Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 18e57d34-480f-4e1e-a4cc-aaed0ed370b6 · inbound

EarlyTom: Early Token Compression Completes Fast Video Understanding cites this paper.

EarlyTom: Early Token Compression Completes Fast Video Understanding Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 54

Resolution: malformed identifier; local_arxiv, observed 2026-06-29T08:23:15.623304Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ae5a6af0-ccec-41c8-99a8-962d15077477 · inbound

Disentangling Semantic Attention from Structural Bias in the Attention Manifold cites this paper.

Disentangling Semantic Attention from Structural Bias in the Attention Manifold Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

Reference 38

Resolution: unresolved; no resolver link, observed 2026-07-31T23:19:56.727294Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.