Paper Citation Record · LEDGER

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries

As of 21 August 2026, this Paper Citation Record lists 43 of 43 outbound references and 2 inbound Pith citation observations for arXiv:2506.12320.

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

pith.paper-citation-record.v1
2506.12320 v1

Coverage vector

measured 43 of 43 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T00:57:29.773071Z

measured 45 of 45 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 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-04T05:55:44.664325Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-16T11:57:50.038881Z

Reference resolution

43 of 43 outbound references displayed

verified exact2
verified fuzzy28
unresolved13
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 453dba99-779b-4672-881e-03b58a4c2268 · outbound

This paper cites Pytorch: An imperative style, high-performance deep learning library,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Pytorch: An imperative style, high-performance deep learning library,

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.754729Z

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 528758d1-dde7-4c2f-b17e-483eb56560b6 · outbound

This paper cites TensorFlow: Large-scale machine learning on heterogeneous systems,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries TensorFlow: Large-scale machine learning on heterogeneous systems,

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.740423Z

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 3d0e62f4-9f2c-4fcb-9ef8-c37ca9dbf5a4 · outbound

This paper cites Transformers: State-of- the-art natural language processing,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Transformers: State-of- the-art natural language processing,

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:27.488140Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 528d814a-29e8-4958-9065-c2759c593807 · outbound

This paper cites Efficient memory management for large language model serving with pagedattention,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Efficient memory management for large language model serving with pagedattention,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:27.559541Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00835e5f-9801-441f-992b-22116f2bdda8 · outbound

This paper cites Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters,

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:27.670249Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e40cafe4-77a9-4022-a82a-48315898b8e1 · outbound

This paper cites Architectural foundations for the large language model infrastructures,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Architectural foundations for the large language model infrastructures,

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.707536Z

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 5917def7-4065-456e-88a7-3ee1ba41bf59 · outbound

This paper cites Efficient Training of Large Language Models on Distributed Infrastructures: A Survey.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Efficient Training of Large Language Models on Distributed Infrastructures: A Survey

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:27.917106Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8dcd5d93-ec13-4730-9aa9-9579b152b524 · outbound

This paper cites Bytedance intern fired for planting malicious code in ai models,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Bytedance intern fired for planting malicious code in ai models,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.693364Z

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 b1d66741-ea0b-4aad-beda-b1c42a7e7418 · outbound

This paper cites Critical bugs which affected everyone’s llm training,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Critical bugs which affected everyone’s llm training,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.678552Z

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 c79dc12e-72d6-46ff-bedf-ac77fcdb1203 · outbound

This paper cites A comprehensive study on deep learning bug characteristics,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries A comprehensive study on deep learning bug characteristics,

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:28.271241Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation df57bc2d-03ee-4ca3-952d-d04c7a1c78b8 · outbound

This paper cites Toward understanding deep learning framework bugs,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Toward understanding deep learning framework bugs,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.665015Z

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 d543a561-601e-46d0-9a34-1b45d6bbb7da · outbound

This paper cites Understanding Performance Problems in Deep Learning Systems.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Understanding Performance Problems in Deep Learning Systems

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-08-07T00:57:30.111772Z

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 ef13d63d-25c0-4f13-bdc1-ff2f5204a92e · outbound

This paper cites Towards understanding the faults of javascript-based deep learning systems,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Towards understanding the faults of javascript-based deep learning systems,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.652064Z

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 3cdbd95b-c5d3-416d-8498-8594cd2d7363 · outbound

This paper cites Large Language Models are Zero-Shot Fuzzers: Fuzzing Deep-Learning Libraries via Large Language Models.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Large Language Models are Zero-Shot Fuzzers: Fuzzing Deep-Learning Libraries via Large Language Models

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:28.748900Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 639bf9e4-662f-441c-8273-ce71fe82157c · outbound

This paper cites Jitfuzz: Coverage-guided fuzzing for jvm just-in-time compilers,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Jitfuzz: Coverage-guided fuzzing for jvm just-in-time compilers,

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:28.855207Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4dec4a12-c1b5-4bad-bce0-984b1a75ed6d · outbound

This paper cites Nnsmith: Generating diverse and valid test cases for deep learning compilers,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Nnsmith: Generating diverse and valid test cases for deep learning compilers,

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:28.983979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 485774b9-61fd-47e3-9d47-1e707fdece1e · outbound

This paper cites A comprehensive study on challenges in deploying deep learning based software,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries A comprehensive study on challenges in deploying deep learning based software,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.638339Z

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 b97d64ff-bb15-49ce-aaf6-8a42962d14db · outbound

This paper cites Our repository,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Our repository,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.625323Z

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 cee74161-4233-4016-b690-75cba9cc4932 · outbound

This paper cites Lora: Low-rank adaptation of large language models.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Lora: Low-rank adaptation of large language models

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:29.147789Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 570be259-7111-4ef7-89ea-8e040183bc54 · outbound

This paper cites Qlora: Efficient finetuning of quantized llms,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Qlora: Efficient finetuning of quantized llms,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:29.206536Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3c3babc-ef4f-435f-b5fa-ca8d9481e112 · outbound

This paper cites Bugs in pods: Understanding bugs in container runtime systems,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Bugs in pods: Understanding bugs in container runtime systems,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.594324Z

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 f9c64b41-c551-4010-b961-e13154c369c9 · outbound

This paper cites Large-scale analysis of non-termination bugs in real-world oss projects,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Large-scale analysis of non-termination bugs in real-world oss projects,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.580570Z

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 647b519f-eb55-416d-9aa8-63e6b0e0c527 · outbound

This paper cites A comprehensive study of deep learning compiler bugs,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries A comprehensive study of deep learning compiler bugs,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.566952Z

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 8e7e0755-7acd-45d4-9eec-5d350ac69f15 · outbound

This paper cites The measurement of observer agreement for categorical data,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries The measurement of observer agreement for categorical data,

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:29.509007Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 87b32193-86ac-4b38-b0ee-e3afc34bab57 · outbound

This paper cites Error happens when using siglipvisionmodel.forward with interpo- late pos encoding=true,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Error happens when using siglipvisionmodel.forward with interpo- late pos encoding=true,

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.544047Z

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 da52f6ad-a19d-4d1e-b3b3-7c4978079b17 · outbound

This paper cites Mistral ’sentencepiecetokenizer’ object has no attribute ’id to byte piece’,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Mistral ’sentencepiecetokenizer’ object has no attribute ’id to byte piece’,

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.531151Z

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 7f3546e7-6098-456f-8969-66e456facb3b · outbound

This paper cites [v1] fix when max model len is not divisible by block size,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries [v1] fix when max model len is not divisible by block size,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.518097Z

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 e60a591f-0d93-4697-b5b1-e1e69f30b586 · outbound

This paper cites Performance bug example in transformers,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Performance bug example in transformers,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.505452Z

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 694129be-6ad8-4190-9ae3-f34f00b03348 · outbound

This paper cites Replace torch.tensor() with torch.from numpy() when process- ing numpy arrays,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Replace torch.tensor() with torch.from numpy() when process- ing numpy arrays,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.492092Z

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 31d40df1-c481-4080-ba7b-dd8cc118f049 · outbound

This paper cites Rename phi-3 rope scaling type,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Rename phi-3 rope scaling type,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.479141Z

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 f63389e3-288d-4391-b50a-6d49f099fb24 · outbound

This paper cites Llava: fix checkpoint doc,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Llava: fix checkpoint doc,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.466043Z

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 b34b20b9-b4d8-488b-bed8-91696145d67e · outbound

This paper cites A survey of deep learning library testing methods,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries A survey of deep learning library testing methods,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.452320Z

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 32fddd53-d6b6-4903-990c-f887698eba94 · outbound

This paper cites Audee: Automated testing for deep learning frameworks,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Audee: Automated testing for deep learning frameworks,

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.438058Z

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 ada50965-31ce-421f-8c89-753153a16517 · outbound

This paper cites Writing unit tests,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Writing unit tests,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.424276Z

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 1c90985d-187a-4f88-973a-0509ae1dd0cb · outbound

This paper cites Fixes for issue #33763 in idefics2 model,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Fixes for issue #33763 in idefics2 model,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.410183Z

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 abb0ab0e-961f-4245-8bb3-78bbab86de8e · outbound

This paper cites Large Language Models are Edge-Case Fuzzers: Testing Deep Learning Libraries via FuzzGPT.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Large Language Models are Edge-Case Fuzzers: Testing Deep Learning Libraries via FuzzGPT

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:29.743159Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d35e5e46-8c8c-4524-862d-8fc27ae2e711 · outbound

This paper cites Silent bugs in deep learning frameworks: an empirical study of keras and tensorflow,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Silent bugs in deep learning frameworks: an empirical study of keras and tensorflow,

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.394500Z

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 8fc3175d-9b54-45c5-a053-9f2d790e7536 · outbound

This paper cites Understanding performance problems in deep learning systems,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Understanding performance problems in deep learning systems,

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.379234Z

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 99da57d2-7a80-4e7a-88ea-983fea784dc9 · outbound

This paper cites To- wards neural architecture-aware exploration of compiler optimizations in a deep learning{graph}compiler,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries To- wards neural architecture-aware exploration of compiler optimizations in a deep learning{graph}compiler,

Reference 39

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.354559Z

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 447b1e4d-a5ff-46b7-a9cd-d59920f29b5f · outbound

This paper cites Moving deep learning into web browser: How far can we go?.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Moving deep learning into web browser: How far can we go?

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.340920Z

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 d837a3dd-2652-4aba-9228-4d3ef75104ea · outbound

This paper cites Beyond accuracy: An empirical study on unit testing in open-source deep learning projects,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Beyond accuracy: An empirical study on unit testing in open-source deep learning projects,

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T00:57:30.326384Z

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 ee9b145d-892d-4afa-9a54-130fa220d607 · outbound

This paper cites The Seeds of the FUTURE Sprout from History: Fuzzing for Unveiling Vulnerabilities in Prospective Deep-Learning Libraries.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries The Seeds of the FUTURE Sprout from History: Fuzzing for Unveiling Vulnerabilities in Prospective Deep-Learning Libraries

Reference 42

Resolution: verified exact; local_arxiv, observed 2026-08-07T00:57:29.907927Z

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 139ad0d1-4b6f-432f-909a-6ff04e7226a8 · outbound

This paper cites Free lunch for testing: fuzzing deep-learning libraries from open source,.

The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries Free lunch for testing: fuzzing deep-learning libraries from open source,

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-07T00:57:29.773071Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 419695e7-26af-4828-9a40-967b3b7bf6d3 · inbound

When Agents Fail: A Comprehensive Study of Bugs in LLM Agents with Automated Labeling cites this paper.

When Agents Fail: A Comprehensive Study of Bugs in LLM Agents with Automated Labeling The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries

Reference 42

Resolution: verified exact; arxiv_id, observed 2026-05-16T11:57:50.040591Z

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 e3e50f99-040c-4681-a17a-014fd494247c · inbound

Real Faults in Model Context Protocol (MCP) Software: a Comprehensive Taxonomy cites this paper.

Real Faults in Model Context Protocol (MCP) Software: a Comprehensive Taxonomy The Foundation Cracks: A Comprehensive Study on Bugs and Testing Practices in LLM Libraries

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-04T05:55:44.664325Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.