Paper Citation Record · LEDGER

Position: The Most Expensive Part of an LLM should be its Training Data

As of 17 August 2026, this Paper Citation Record lists 48 of 48 outbound references and 5 inbound Pith citation observations for arXiv:2504.12427.

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

pith.paper-citation-record.v1
2504.12427 v1

Coverage vector

measured 48 of 48 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-16T12:34:53.589828Z

measured 53 of 53 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-16T06:30:59.297886+00:00

measured 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T10:29:44.273599Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-04T11:39:46.610216Z

Reference resolution

48 of 48 outbound references displayed

verified exact1
verified fuzzy20
unresolved27
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 6ffee95a-85f2-44e2-8647-575ce81577e8 · outbound

This paper cites https://huggingface.co/collections/r-three/common-pile-665a13e48528df6b00416dc0, 2025.

Position: The Most Expensive Part of an LLM should be its Training Data https://huggingface.co/collections/r-three/common-pile-665a13e48528df6b00416dc0, 2025

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.222109Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation c0836bd0-436d-4232-a7b5-5f1de43df576 · outbound

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

Position: The Most Expensive Part of an LLM should be its Training Data Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.420387Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29a1ac42-8e9a-48f9-bc48-2966b1f50e70 · outbound

This paper cites Phi-4 Technical Report.

Position: The Most Expensive Part of an LLM should be its Training Data Phi-4 Technical Report

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.424961Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94a2aa78-8802-47da-b446-a0bfd1abc6e2 · outbound

This paper cites Alden newspapers v.

Position: The Most Expensive Part of an LLM should be its Training Data Alden newspapers v

Reference 4

Resolution: verified exact; raw_fallback, observed 2026-08-16T12:34:53.976485Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 60d5f8ad-9b5f-4da0-bfcd-50cdfb622b42 · outbound

This paper cites M., and Weber, G.

Position: The Most Expensive Part of an LLM should be its Training Data M., and Weber, G

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.433029Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7adfd76b-b4c3-45f7-b7cc-788c976aecab · outbound

This paper cites Authors guild v.

Position: The Most Expensive Part of an LLM should be its Training Data Authors guild v

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.206262Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 72bb5d94-82ea-4e6d-abb9-10b563348510 · outbound

This paper cites What is Your Data Worth to GPT? LLM-Scale Data Valuation with Influence Functions.

Position: The Most Expensive Part of an LLM should be its Training Data What is Your Data Worth to GPT? LLM-Scale Data Valuation with Influence Functions

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.441434Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 558c1ceb-306d-4caa-86a9-1ef6e5f74c7d · outbound

This paper cites Common crawl dataset.

Position: The Most Expensive Part of an LLM should be its Training Data Common crawl dataset

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.196771Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 3518d52d-387a-4a8d-9541-f23ebcd8f2f9 · outbound

This paper cites Concord music group v.

Position: The Most Expensive Part of an LLM should be its Training Data Concord music group v

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.187530Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 252eac39-7388-4717-be9e-6a70a8dc6d8e · outbound

This paper cites The rising costs of training frontier AI models.

Position: The Most Expensive Part of an LLM should be its Training Data The rising costs of training frontier AI models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.452722Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cbefb914-e2ca-4d11-93c0-b8d709d75485 · outbound

This paper cites Ai is a lot of work.

Position: The Most Expensive Part of an LLM should be its Training Data Ai is a lot of work

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.177432Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 0b9777a9-53d8-4c17-988e-4a47343008d5 · outbound

This paper cites Encyclopedia britannica 15th edition.

Position: The Most Expensive Part of an LLM should be its Training Data Encyclopedia britannica 15th edition

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.167706Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 9ac7fb88-d47c-487b-9be8-f0936f3ab916 · outbound

This paper cites Data on notable ai models, 2024.

Position: The Most Expensive Part of an LLM should be its Training Data Data on notable ai models, 2024

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.157863Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation c957d152-6d94-4247-b56b-21a7ecc89e37 · outbound

This paper cites DataComp: In search of the next generation of multimodal datasets.

Position: The Most Expensive Part of an LLM should be its Training Data DataComp: In search of the next generation of multimodal datasets

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.468171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01f09962-6c0c-46ab-a7ef-e7905f4ce083 · outbound

This paper cites Language models scale reliably with over-training and on downstream tasks.

Position: The Most Expensive Part of an LLM should be its Training Data Language models scale reliably with over-training and on downstream tasks

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.471844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cba7a639-6920-48ad-a543-dbde1f5b53a7 · outbound

This paper cites Data Shapley: Equitable Valuation of Data for Machine Learning.

Position: The Most Expensive Part of an LLM should be its Training Data Data Shapley: Equitable Valuation of Data for Machine Learning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.475484Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 03d72fee-497e-469b-a1cd-5e46266b1faa · outbound

This paper cites Evaluation of Similarity-based Explanations.

Position: The Most Expensive Part of an LLM should be its Training Data Evaluation of Similarity-based Explanations

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.479173Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95e6842e-d01d-40fb-95c2-875558897655 · outbound

This paper cites Training Compute-Optimal Large Language Models.

Position: The Most Expensive Part of an LLM should be its Training Data Training Compute-Optimal Large Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.482888Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e0e7818b-ac76-4f46-a3a7-5c125e939bf3 · outbound

This paper cites Statistics on wages.

Position: The Most Expensive Part of an LLM should be its Training Data Statistics on wages

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.147863Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 6bad627c-5881-4537-af10-2065619543c6 · outbound

This paper cites Scaling Laws for Neural Language Models.

Position: The Most Expensive Part of an LLM should be its Training Data Scaling Laws for Neural Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.490580Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b779aefe-5e3e-4bf2-9355-d243a13effb7 · outbound

This paper cites Understanding Black-box Predictions via Influence Functions.

Position: The Most Expensive Part of an LLM should be its Training Data Understanding Black-box Predictions via Influence Functions

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.494514Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dc20d4d5-255e-42cc-80ea-f82e155d7c85 · outbound

This paper cites OpenAssistant Conversations -- Democratizing Large Language Model Alignment.

Position: The Most Expensive Part of an LLM should be its Training Data OpenAssistant Conversations -- Democratizing Large Language Model Alignment

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.498197Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2d526e30-7940-40ee-9d3b-5bf491476136 · outbound

This paper cites Releasing Common Corpus: the largest public domain dataset for training LLMs.

Position: The Most Expensive Part of an LLM should be its Training Data Releasing Common Corpus: the largest public domain dataset for training LLMs

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.138207Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation b173a557-7554-43ba-9c09-4b38c6b0d91f · outbound

This paper cites DataComp-LM: In search of the next generation of training sets for language models.

Position: The Most Expensive Part of an LLM should be its Training Data DataComp-LM: In search of the next generation of training sets for language models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.505422Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 589d0d14-d539-4794-a32c-62452bcf0ab2 · outbound

This paper cites Consent in Crisis: The Rapid Decline of the AI Data Commons.

Position: The Most Expensive Part of an LLM should be its Training Data Consent in Crisis: The Rapid Decline of the AI Data Commons

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.509363Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9175506-313c-4a0e-88e3-645e166e0898 · outbound

This paper cites SILO Language Models: Isolating Legal Risk In a Nonparametric Datastore.

Position: The Most Expensive Part of an LLM should be its Training Data SILO Language Models: Isolating Legal Risk In a Nonparametric Datastore

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.513176Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50021800-f7b3-4840-8bfa-76e96383ddf3 · outbound

This paper cites AI models that cost \ 1 billion to train are underway, \ 100 billion models coming — largest current models take “only” \ 100 million to train: Anthropic CEO.

Position: The Most Expensive Part of an LLM should be its Training Data AI models that cost \ 1 billion to train are underway, \ 100 billion models coming — largest current models take “only” \ 100 million to train: Anthropic CEO

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.128186Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 34d09960-edb5-43cd-87e2-dfe3162eaac7 · outbound

This paper cites Scaling Data-Constrained Language Models.

Position: The Most Expensive Part of an LLM should be its Training Data Scaling Data-Constrained Language Models

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.520465Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 51e8f85b-56e4-42ca-806b-f3d77227f000 · outbound

This paper cites New york times v.

Position: The Most Expensive Part of an LLM should be its Training Data New york times v

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.117493Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 42b77bc3-054d-4116-bc9e-b719feca545f · outbound

This paper cites TRAK: Attributing Model Behavior at Scale.

Position: The Most Expensive Part of an LLM should be its Training Data TRAK: Attributing Model Behavior at Scale

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.527627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c16cddf-21d2-4a9b-abad-7c7caead9e09 · outbound

This paper cites an unresolved cited work.

Position: The Most Expensive Part of an LLM should be its Training Data Unresolved cited work

Reference 31

Resolution: unresolved; raw_fallback, observed 2026-08-16T12:34:54.107235Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 4e90d540-21fa-4d3f-a022-c765ce57d782 · outbound

This paper cites Multiple ai companies bypassing web standard to scrape publisher sites, licensing firm says.

Position: The Most Expensive Part of an LLM should be its Training Data Multiple ai companies bypassing web standard to scrape publisher sites, licensing firm says

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.096999Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 6bc0e423-6cf3-4234-9b54-9de913a1b9e9 · outbound

This paper cites The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale.

Position: The Most Expensive Part of an LLM should be its Training Data The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.538809Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 117ed8c8-2b11-4ca1-9282-57fa9ad774a7 · outbound

This paper cites Estimating Training Data Influence by Tracing Gradient Descent.

Position: The Most Expensive Part of an LLM should be its Training Data Estimating Training Data Influence by Tracing Gradient Descent

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.542296Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8f50bfd0-16cd-479b-85d2-b1164801cce4 · outbound

This paper cites Reddit and openai build partnership.

Position: The Most Expensive Part of an LLM should be its Training Data Reddit and openai build partnership

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.086656Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation a27c7275-92b1-4929-a0ce-de699e929c1d · outbound

This paper cites Thomson reuters' adjusted eps beats expectations, ai boosts results.

Position: The Most Expensive Part of an LLM should be its Training Data Thomson reuters' adjusted eps beats expectations, ai boosts results

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.075904Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 22b3cdde-bdef-4cd2-b62d-524a157ec2fd · outbound

This paper cites Shutterstock expands partnership with openai, signs new six-year agreement to provide high-quality training data.

Position: The Most Expensive Part of an LLM should be its Training Data Shutterstock expands partnership with openai, signs new six-year agreement to provide high-quality training data

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.064402Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 56e88207-67cb-4cda-b630-7c7928ad4f1e · outbound

This paper cites Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research.

Position: The Most Expensive Part of an LLM should be its Training Data Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.556356Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5c5e27f6-786f-482f-8476-120d83c243b4 · outbound

This paper cites The atlantic announces product and content partnership with openai.

Position: The Most Expensive Part of an LLM should be its Training Data The atlantic announces product and content partnership with openai

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.052727Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 02da8847-5bea-46ec-90e4-ab9716fc5df4 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Position: The Most Expensive Part of an LLM should be its Training Data LLaMA: Open and Efficient Foundation Language Models

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.563650Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dde8342c-841c-44e7-9816-8d01bed89acc · outbound

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

Position: The Most Expensive Part of an LLM should be its Training Data Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.567083Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cbf0bcf5-7040-43f4-ae36-73e83a968d80 · outbound

This paper cites Will we run out of data? Limits of LLM scaling based on human-generated data.

Position: The Most Expensive Part of an LLM should be its Training Data Will we run out of data? Limits of LLM scaling based on human-generated data

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.570304Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d74ac334-eff0-41e9-a858-6d7d0eb6f15b · outbound

This paper cites Vox media and openai form strategic content and product partnership.

Position: The Most Expensive Part of an LLM should be its Training Data Vox media and openai form strategic content and product partnership

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.042052Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 3b58b3df-6705-40f0-bd7c-321f26fd55e5 · outbound

This paper cites Html standard.

Position: The Most Expensive Part of an LLM should be its Training Data Html standard

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.030794Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 78226ba2-210d-4061-9217-5374472983b3 · outbound

This paper cites Call for Papers -- The BabyLM Challenge: Sample-efficient pretraining on a developmentally plausible corpus.

Position: The Most Expensive Part of an LLM should be its Training Data Call for Papers -- The BabyLM Challenge: Sample-efficient pretraining on a developmentally plausible corpus

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.580062Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d9761ef1-9d90-41fb-b3ac-18d1a352da19 · outbound

This paper cites Breaking down emerging segments in the ai content licensing landscape.

Position: The Most Expensive Part of an LLM should be its Training Data Breaking down emerging segments in the ai content licensing landscape

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.019864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e321c3e6-f2cd-44a7-b3d7-749b33eb06b1 · outbound

This paper cites Articulating value from data.

Position: The Most Expensive Part of an LLM should be its Training Data Articulating value from data

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-16T12:34:54.007464Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation d301cb7b-75c5-43ff-bab2-da3c74223efc · outbound

This paper cites WildChat: 1M ChatGPT Interaction Logs in the Wild.

Position: The Most Expensive Part of an LLM should be its Training Data WildChat: 1M ChatGPT Interaction Logs in the Wild

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-16T12:34:53.589828Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 179fac02-40ff-4f0c-a983-85f16f6422af · inbound

The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text cites this paper.

The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text Position: The Most Expensive Part of an LLM should be its Training Data

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-07T10:29:44.273599Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7895a035-1447-469b-a995-8b83e5474abe · inbound

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation cites this paper.

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation Position: The Most Expensive Part of an LLM should be its Training Data

Reference 34

Resolution: metadata mismatch; arxiv_id, observed 2026-05-15T14:25:55.503750Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation de189bdf-7b49-496d-8753-7ab89b8483e5 · inbound

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation cites this paper.

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation Position: The Most Expensive Part of an LLM should be its Training Data

Reference 34

Resolution: unresolved; no resolver link, observed 2026-07-15T12:17:47.123519Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2438a06-593e-4df0-9cc9-bd8ff63152eb · inbound

On the Fragility of Data Attribution When Learning Is Distributed cites this paper.

On the Fragility of Data Attribution When Learning Is Distributed Position: The Most Expensive Part of an LLM should be its Training Data

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-19T15:17:39.452630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation fc6c92ef-bb74-4ea7-8367-1d7781beffa3 · inbound

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation cites this paper.

FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation Position: The Most Expensive Part of an LLM should be its Training Data

Reference 34

Resolution: verified exact; arxiv_id, observed 2026-07-04T11:39:46.611909Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.