Paper Citation Record · LEDGER

PowLU: An Activation Function for Stable Pre-Training of LLMs

As of 14 August 2026, this Paper Citation Record lists 26 of 26 outbound references and 1 inbound Pith citation observation for arXiv:2605.25704.

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

pith.paper-citation-record.v1
2605.25704 v1

Coverage vector

measured 26 of 26 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-29T21:48:45.339523Z

measured 27 of 27 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-10T10:23:35.898298Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-10T10:23:36.660473Z

Reference resolution

26 of 26 outbound references displayed

verified exact16
verified fuzzy0
unresolved4
parse uncertain0
malformed identifier0
metadata mismatch6

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 5b42babe-5cbf-4f82-8751-6714677a9493 · outbound

This paper cites gpt-oss-120b & gpt-oss-20b Model Card.

PowLU: An Activation Function for Stable Pre-Training of LLMs gpt-oss-120b & gpt-oss-20b Model Card

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.476830Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 42448976-1437-4ce1-9bbd-ec53f81a92bf · outbound

This paper cites Faster and Lighter LLMs: A Survey on Current Challenges and Way Forward.

PowLU: An Activation Function for Stable Pre-Training of LLMs Faster and Lighter LLMs: A Survey on Current Challenges and Way Forward

Reference 2

Resolution: verified exact; arxiv_id, observed 2026-06-29T21:53:59.481926Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation dbaed6f4-b419-4489-952c-0c9affabf346 · outbound

This paper cites Evaluating Large Language Models Trained on Code.

PowLU: An Activation Function for Stable Pre-Training of LLMs Evaluating Large Language Models Trained on Code

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.479326Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d4f90ef5-c362-46b0-bd4e-0cb07aeba036 · outbound

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

PowLU: An Activation Function for Stable Pre-Training of LLMs Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.438177Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation efaa5594-4ab2-4459-b950-8688c142f593 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

PowLU: An Activation Function for Stable Pre-Training of LLMs Training Verifiers to Solve Math Word Problems

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.471374Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e7f7d895-ffaf-4283-a7d4-a227224789cc · outbound

This paper cites The Llama 3 Herd of Models.

PowLU: An Activation Function for Stable Pre-Training of LLMs The Llama 3 Herd of Models

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.437780Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c77a3e84-30d4-4fc4-a870-dbab44402c3f · outbound

This paper cites Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities.

PowLU: An Activation Function for Stable Pre-Training of LLMs Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-07-30T01:18:51.459998Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 7c7e6766-8e70-4a3c-8084-955eebe8e3c4 · outbound

This paper cites Gaussian Error Linear Units (GELUs).

PowLU: An Activation Function for Stable Pre-Training of LLMs Gaussian Error Linear Units (GELUs)

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.457829Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c6e894d6-a82c-47de-b36e-95c05ee6a2f3 · outbound

This paper cites Measuring Mathematical Problem Solving With the MATH Dataset.

PowLU: An Activation Function for Stable Pre-Training of LLMs Measuring Mathematical Problem Solving With the MATH Dataset

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.469094Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 5004ce4c-b045-4b48-bbd4-71c1b3c0c864 · outbound

This paper cites Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks.

PowLU: An Activation Function for Stable Pre-Training of LLMs Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks

Reference 10

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T21:53:59.469206Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 471e4a6d-6f7b-42bf-b03e-5d2c5a83193f · outbound

This paper cites To FP8 and Back Again: Quantifying Reduced Precision Effects on LLM Training Stability.

PowLU: An Activation Function for Stable Pre-Training of LLMs To FP8 and Back Again: Quantifying Reduced Precision Effects on LLM Training Stability

Reference 11

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T21:53:59.474668Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 3ed4298d-01bc-423a-88dc-360e26bb7114 · outbound

This paper cites CMMLU: Measuring massive multitask language understanding in Chinese.

PowLU: An Activation Function for Stable Pre-Training of LLMs CMMLU: Measuring massive multitask language understanding in Chinese

Reference 12

Resolution: unresolved; no resolver link, observed 2026-06-29T21:48:45.339523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25dcb5b0-b353-42e2-a155-09eab156e9f5 · outbound

This paper cites Large Language Models: A Survey.

PowLU: An Activation Function for Stable Pre-Training of LLMs Large Language Models: A Survey

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.450101Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation d0cf46d7-a057-4c14-ab32-228a08bc9970 · outbound

This paper cites ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models.

PowLU: An Activation Function for Stable Pre-Training of LLMs ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-06-29T21:53:59.458769Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation c2fef1fe-201a-4872-9993-551dc61a4c9c · outbound

This paper cites A Theory on Adam Instability in Large-Scale Machine Learning.

PowLU: An Activation Function for Stable Pre-Training of LLMs A Theory on Adam Instability in Large-Scale Machine Learning

Reference 15

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T21:53:59.472017Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation f1c529a9-67c1-4cd0-a57b-64ec992dfc97 · outbound

This paper cites Searching for Activation Functions.

PowLU: An Activation Function for Stable Pre-Training of LLMs Searching for Activation Functions

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.446496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation db827b29-b885-4996-a4c7-d6d78b901052 · outbound

This paper cites GLU Variants Improve Transformer.

PowLU: An Activation Function for Stable Pre-Training of LLMs GLU Variants Improve Transformer

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.463432Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 1f66a506-09a7-4e62-8088-fc7b61845d2e · outbound

This paper cites OpenAI GPT-5 System Card.

PowLU: An Activation Function for Stable Pre-Training of LLMs OpenAI GPT-5 System Card

Reference 18

Resolution: metadata mismatch; local_arxiv, observed 2026-06-29T21:53:59.461195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 6b15d205-0aa8-49a1-9b95-e488aff291d6 · outbound

This paper cites Challenging big-bench tasks and whether chain-of-thought can solve them.

PowLU: An Activation Function for Stable Pre-Training of LLMs Challenging big-bench tasks and whether chain-of-thought can solve them

Reference 19

Resolution: unresolved; no resolver link, observed 2026-06-29T21:48:45.339523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7997d0d8-8c0f-460d-95c3-99fb95c84043 · outbound

This paper cites Gemini: A Family of Highly Capable Multimodal Models.

PowLU: An Activation Function for Stable Pre-Training of LLMs Gemini: A Family of Highly Capable Multimodal Models

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.443807Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation e491433f-a1ca-4ed9-9b7e-dc546b46bf81 · outbound

This paper cites Attention Residuals.

PowLU: An Activation Function for Stable Pre-Training of LLMs Attention Residuals

Reference 21

Resolution: verified exact; local_arxiv, observed 2026-06-29T21:53:59.435252Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 8f4a9196-1d4e-47b8-89b7-db6d0d6d87d1 · outbound

This paper cites Every activa- tion boosted: Scaling general reasoner to 1 trillion open language foundation.arXiv preprint arXiv:2510.22115.

PowLU: An Activation Function for Stable Pre-Training of LLMs Every activa- tion boosted: Scaling general reasoner to 1 trillion open language foundation.arXiv preprint arXiv:2510.22115

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-06-29T21:53:59.440597Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 21f7cf64-099b-4eac-8aab-6f001aa74d47 · outbound

This paper cites CMATH: Can Your Language Model Pass Chinese Elementary School Math Test?.

PowLU: An Activation Function for Stable Pre-Training of LLMs CMATH: Can Your Language Model Pass Chinese Elementary School Math Test?

Reference 23

Resolution: metadata mismatch; arxiv_id, observed 2026-06-29T21:53:59.441133Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 3b14062b-14ec-4365-90b5-0fa4e13d88de · outbound

This paper cites Qwen3 Technical Report.

PowLU: An Activation Function for Stable Pre-Training of LLMs Qwen3 Technical Report

Reference 24

Resolution: metadata mismatch; local_arxiv, observed 2026-06-29T21:53:59.430000Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

Observation 922cc6e7-5a69-44f0-aad2-d50deea19c9c · outbound

This paper cites Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, and Nan Duan.

PowLU: An Activation Function for Stable Pre-Training of LLMs Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, and Nan Duan

Reference 25

Resolution: unresolved; no resolver link, observed 2026-06-29T21:48:45.339523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66b10ee3-4a4b-49cb-ae3d-582fffc5f90f · outbound

This paper cites The sigmoid function is non-linear (Ramapuram et al., 2025).

PowLU: An Activation Function for Stable Pre-Training of LLMs The sigmoid function is non-linear (Ramapuram et al., 2025)

Reference 26

Resolution: unresolved; no resolver link, observed 2026-06-29T21:48:45.339523Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 2e625571-b393-4467-a913-a10ad4133e08 · inbound

Is SwiGLU's Open Positive Tail Necessary? Evidence from Closed-Tail Gating with MemGLU cites this paper.

Is SwiGLU's Open Positive Tail Necessary? Evidence from Closed-Tail Gating with MemGLU PowLU: An Activation Function for Stable Pre-Training of LLMs

Reference 2026

Resolution: metadata mismatch; local_arxiv, observed 2026-08-10T10:23:36.665046Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.