Paper Citation Record · LEDGER

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs

As of 21 August 2026, this Paper Citation Record lists 32 of 32 outbound references and 3 inbound Pith citation observations for arXiv:2505.02009.

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

pith.paper-citation-record.v1
2505.02009 v3

Coverage vector

measured 32 of 32 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T04:09:26.124106Z

measured 35 of 35 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-21T06:32:19.484+00:00

measured 3 of 3 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T23:13:05.224677Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-02T12:26:57.033442Z

Reference resolution

32 of 32 outbound references displayed

verified exact0
verified fuzzy22
unresolved10
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation d4925574-3d7e-4987-9230-c4838ca40804 · outbound

This paper cites Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:25.966440Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05346a7b-52ee-4d55-b419-aa249f7f858e · outbound

This paper cites Sparks of Artificial General Intelligence: Early experiments with GPT-4.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Sparks of Artificial General Intelligence: Early experiments with GPT-4

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:25.994378Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3b647bc3-f328-4af7-9a8e-9a8141c1cba4 · outbound

This paper cites Illicit darkweb clas- sification via natural-language processing: Classifying illicit content of webpages based on textual information.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Illicit darkweb clas- sification via natural-language processing: Classifying illicit content of webpages based on textual information

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.660624Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f6130c86-cf74-411b-9289-054db11cb7ac · outbound

This paper cites HateBERT: Retrain- ing BERT for abusive language detection in English.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs HateBERT: Retrain- ing BERT for abusive language detection in English

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.641747Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 345d4b0e-9013-4a49-8489-fb6b75a02e73 · outbound

This paper cites an unresolved cited work.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-16T04:09:26.602751Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b5088d15-c80a-4efe-bc58-362ccaa25ea7 · outbound

This paper cites The Llama 3 Herd of Models.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs The Llama 3 Herd of Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:26.024807Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 699ffc2a-35f1-4296-8b27-78c90485a0d9 · outbound

This paper cites an unresolved cited work.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Unresolved cited work

Reference 13

Resolution: unresolved; raw_fallback, observed 2026-08-16T04:09:26.568895Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a215cf9b-a5b3-4ed5-a531-9849e4c774f4 · outbound

This paper cites [Gemma, 2024] Riviere et al.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Gemma, 2024] Riviere et al

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.552504Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation dc7efd2c-c8cd-438f-a6dc-3f25025189ba · outbound

This paper cites A survey on automated fact-checking.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs A survey on automated fact-checking

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:26.038331Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7742323f-cfe7-45af-b66b-25d55f925ff4 · outbound

This paper cites Llama guard: Llm-based input-output safeguard for human-ai conversations,.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Llama guard: Llm-based input-output safeguard for human-ai conversations,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.524585Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e413571b-7068-4038-84e7-d96ac011f2d8 · outbound

This paper cites Perplexed by quality: A perplexity-based method for adult and harmful content detection in multilingual heterogeneous web data,.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Perplexed by quality: A perplexity-based method for adult and harmful content detection in multilingual heterogeneous web data,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.506568Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d25c268e-e2b4-4d1a-8ebd-aa952b333597 · outbound

This paper cites A pretrainer’s guide to training data: Measuring the effects of data age, domain coverage, quality, & toxicity.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs A pretrainer’s guide to training data: Measuring the effects of data age, domain coverage, quality, & toxicity

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.486439Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5a8a4753-8141-4d5b-8b30-b5d5b6844b03 · outbound

This paper cites [Loshchilov and Hutter, 2019] Ilya Loshchilov and Frank Hutter.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Loshchilov and Hutter, 2019] Ilya Loshchilov and Frank Hutter

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.468220Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c2b61ccc-d146-422d-9398-fd8766926cba · outbound

This paper cites [Markov et al., 2023] Todor Markov, Chong Zhang, Sand- hini Agarwal, Florentine Eloundou Nekoul, Theodore Lee, Steven Adler, Angela Jiang, and Lilian Weng.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Markov et al., 2023] Todor Markov, Chong Zhang, Sand- hini Agarwal, Florentine Eloundou Nekoul, Theodore Lee, Steven Adler, Angela Jiang, and Lilian Weng

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.432060Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 51e11f62-3248-474f-be49-7c79497feb97 · outbound

This paper cites Llama 3.2: Revolutionizing edge ai and vision with open, customizable models.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Llama 3.2: Revolutionizing edge ai and vision with open, customizable models

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.412569Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a8065ae7-ba00-41c9-8ae8-fb8d563ae6cd · outbound

This paper cites Mistral-7b-v0.3.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Mistral-7b-v0.3

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.396544Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 22d0562e-55fb-4037-a5d2-c1152b222359 · outbound

This paper cites Ollama - get up and running with large language models.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Ollama - get up and running with large language models

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.379148Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 1fc408a3-7f32-4ab1-8109-72b8d66f5fbc · outbound

This paper cites Data, data everywhere: A guide for pretraining dataset construction.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Data, data everywhere: A guide for pretraining dataset construction

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.361071Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4635636d-4916-40c5-bc2a-c5e95dae7338 · outbound

This paper cites [Penedo et al., 2024] Guilherme Penedo, Hynek Kydl ´ıˇcek, Loubna Ben allal, Anton Lozhkov, Margaret Mitchell, Colin Raffel, Leandro V on Werra, and Thomas Wolf.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Penedo et al., 2024] Guilherme Penedo, Hynek Kydl ´ıˇcek, Loubna Ben allal, Anton Lozhkov, Margaret Mitchell, Colin Raffel, Leandro V on Werra, and Thomas Wolf

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.343610Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation d53124c4-172b-4a91-b5ec-749c2587d68c · outbound

This paper cites an unresolved cited work.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-08-16T04:09:26.324416Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 45f3d535-32df-4deb-826b-cef15983f794 · outbound

This paper cites Common crawl – building an open web-scale crawl using hadoop,.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Common crawl – building an open web-scale crawl using hadoop,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.305902Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 5a2cb45f-01e1-43db-bdae-8e61c63d6046 · outbound

This paper cites [Truic˘a and Apostol, 2023] Ciprian-Octavian Truic ˘a and Elena-Simona Apostol.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Truic˘a and Apostol, 2023] Ciprian-Octavian Truic ˘a and Elena-Simona Apostol

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.269867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4e8a4e84-1bf9-4403-8acb-ee9130a06e32 · outbound

This paper cites Gomez, Łukasz Kaiser, and Illia Polosukhin.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Gomez, Łukasz Kaiser, and Illia Polosukhin

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.250492Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 0340ade0-f2c9-44f9-ab90-2b20cd62a7fc · outbound

This paper cites Problematic webpage identification: A trilogy of hatespeech, search engines and GPT.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Problematic webpage identification: A trilogy of hatespeech, search engines and GPT

Reference 2010

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.288078Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c2cdef70-35aa-41d9-8261-c059ce7e61ac · outbound

This paper cites [Wei et al., 2022] Jason Wei, Xuezhi Wang, Dale Schuur- mans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs [Wei et al., 2022] Jason Wei, Xuezhi Wang, Dale Schuur- mans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H

Reference 2017

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.233298Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6be51992-5f56-430c-b6b1-998cad748a1d · outbound

This paper cites What’s in the box? an analysis of undesirable content in the Common Crawl corpus.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs What’s in the box? an analysis of undesirable content in the Common Crawl corpus

Reference 2019

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.449968Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 59f5f5cf-870c-4cf2-858a-e0daee52ffc1 · outbound

This paper cites Language models are few-shot learn- ers.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Language models are few-shot learn- ers

Reference 2020

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:25.988757Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da30a9a9-46b4-42e5-bc0f-8e9dc8f1b8af · outbound

This paper cites an unresolved cited work.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Unresolved cited work

Reference 2021

Resolution: unresolved; raw_fallback, observed 2026-08-16T04:09:26.621617Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 2686cbdf-9c57-41ab-8cde-05653776f2ab · outbound

This paper cites GPT-4 Technical Report.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs GPT-4 Technical Report

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-16T04:09:25.977828Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea172620-5d56-45af-95b4-9d8c8b7f8108 · outbound

This paper cites Peters, and Ar- man Cohan.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Peters, and Ar- man Cohan

Reference 2023

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.690671Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4faf27d3-69a6-4065-acf3-ca02546ee48e · outbound

This paper cites Suicidal ideation detection on social me- dia: A review of machine learning methods,.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Suicidal ideation detection on social me- dia: A review of machine learning methods,

Reference 2024

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.708853Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 7de96306-a4bf-4e67-bb0a-fed1115cc866 · outbound

This paper cites Documenting large webtext corpora: A case study on the colossal clean crawled corpus.

Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs Documenting large webtext corpora: A case study on the colossal clean crawled corpus

Reference 2025

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T04:09:26.585711Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Pith citing papers

Observation 803bed5d-fdd5-462f-86f8-adeff192f10d · inbound

Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM cites this paper.

Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs

Reference 212

Resolution: unresolved; no resolver link, observed 2026-08-05T23:13:05.224677Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb034246-6af4-4a59-b1e4-b960e64aae42 · inbound

Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection cites this paper.

Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-05T14:57:12.878381Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea22ca27-f0da-4ec6-8833-486a85e51208 · inbound

Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails cites this paper.

Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails Towards Safer Pretraining: Analyzing and Filtering Harmful Content in Webscale datasets for Responsible LLMs

Reference 23

Resolution: verified exact; arxiv_id, observed 2026-07-02T12:26:57.034737Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.