Paper Citation Record · LEDGER

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function

As of 9 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 0 inbound Pith citation observations for arXiv:2509.01874.

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

pith.paper-citation-record.v1
2509.01874 v1

Coverage vector

measured 20 of 20 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-05T12:12:58.711418Z

measured 20 of 20 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 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

20 of 20 outbound references displayed

verified exact1
verified fuzzy2
unresolved16
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 752b864b-dd3b-4a2e-9d4e-d05b2c7b38e9 · outbound

This paper cites A technical note on bilinear layers for interpretability.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function A technical note on bilinear layers for interpretability

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.642723Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad7e6201-d9e9-4afd-ba6a-0f19ae268d8a · outbound

This paper cites Bilinear MLPs enable weight-based mechanistic interpretability.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Bilinear MLPs enable weight-based mechanistic interpretability

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.646799Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73aacce0-2d64-4817-a30d-52d9c5f3ff2f · outbound

This paper cites Language Modeling with Gated Convolutional Networks.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Language Modeling with Gated Convolutional Networks

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.650015Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5c3dc69c-aa2b-4c3e-86e6-a6b489b7d845 · outbound

This paper cites Gaussian Error Linear Units (GELUs).

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Gaussian Error Linear Units (GELUs)

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.653376Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 541d292b-908b-4409-a1e8-68391be73d9f · outbound

This paper cites Deep Learning using Rectified Linear Units (ReLU).

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Deep Learning using Rectified Linear Units (ReLU)

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.656758Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21dbb8e1-dd4d-42ec-a88a-171ce2664282 · outbound

This paper cites The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web].

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function The MNIST Database of Handwritten Digit Images for Machine Learning Research [Best of the Web]

Reference 6

Resolution: metadata mismatch; raw_fallback, observed 2026-08-05T12:12:59.004140Z

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.

Observation 578ec811-0b78-4f7d-ad07-8c7d47613498 · outbound

This paper cites Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.664289Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3c08d88e-9529-4ed6-afac-efcfde51d616 · outbound

This paper cites TinyStories: How Small Can Language Models Be and Still Speak Coherent English?.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function TinyStories: How Small Can Language Models Be and Still Speak Coherent English?

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.668259Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6a6db192-3788-4001-8934-f1fb85d2bcdd · outbound

This paper cites Sparse Autoencoders Find Highly Interpretable Features in Language Models.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.671602Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6acf53b-6c55-4d41-a815-51dafb103247 · outbound

This paper cites Searching for Activation Functions.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Searching for Activation Functions

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.675631Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9a780023-ffd5-4b56-9fd7-13a82ae898d9 · outbound

This paper cites GLU Variants Improve Transformer.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function GLU Variants Improve Transformer

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.678663Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3a04fe4-25cc-470c-a9d0-a343e44fa7ab · outbound

This paper cites Scaling and evaluating sparse autoencoders.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Scaling and evaluating sparse autoencoders

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.682543Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 008afaf1-7a22-46c6-99bf-91b873aba1aa · outbound

This paper cites Interpreting Attention Layer Outputs with Sparse Autoencoders.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Interpreting Attention Layer Outputs with Sparse Autoencoders

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.685588Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8947d59a-f187-4de5-8f37-06c53b67e5f7 · outbound

This paper cites Modular Networks: Learning to Decompose Neural Computation.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Modular Networks: Learning to Decompose Neural Computation

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-08-05T12:12:58.786863Z

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.

Observation 870149c1-9c5b-4aa0-9b4b-6e85d3b0ba28 · outbound

This paper cites Gradient Routing: Masking Gradients to Localize Computation in Neural Networks.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Gradient Routing: Masking Gradients to Localize Computation in Neural Networks

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T12:12:59.078107Z

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.

Observation e146e084-f787-4fe8-b48a-7981c15b9eb3 · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.699892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a12c618e-8146-4214-92dd-ddbffdc23382 · outbound

This paper cites Decoupled Weight Decay Regularization.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Decoupled Weight Decay Regularization

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.704262Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 764d59c3-7f91-483b-bf80-5fa23d2f59d3 · outbound

This paper cites SGDR: Stochastic Gradient Descent with Warm Restarts.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function SGDR: Stochastic Gradient Descent with Warm Restarts

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-05T12:12:59.067285Z

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.

Observation dfe6cdd7-7ce0-4489-a336-6496dcd1e647 · outbound

This paper cites SGDR: Stochastic Gradient Descent with Warm Restarts.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function SGDR: Stochastic Gradient Descent with Warm Restarts

Reference 2017

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.711418Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a36389f-90a5-4efb-a588-0b7c7eb1214c · outbound

This paper cites Gradient Routing: Masking Gradients to Localize Computation in Neural Networks.

Preserving Bilinear Weight Spectra with a Signed and Shrunk Quadratic Activation Function Gradient Routing: Masking Gradients to Localize Computation in Neural Networks

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-05T12:12:58.695316Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.