Paper Citation Record · LEDGER

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training

As of 17 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 1 inbound Pith citation observation for arXiv:2509.04796.

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

pith.paper-citation-record.v1
2509.04796 v1

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-15T16:33:32.575974Z

measured 25 of 25 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-21T05:46:07.176408Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-21T05:49:41.094772Z

Reference resolution

24 of 24 outbound references displayed

verified exact0
verified fuzzy2
unresolved21
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 9e84c5c2-1ea0-49dc-9e46-c5f9e7c35d8f · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 1

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.248631Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1024d491-7ca7-4d35-b713-16a1fb942ee1 · outbound

This paper cites Strong Model Collapse.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Strong Model Collapse

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:31.909316Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b520443c-eaa0-4d3d-a8a6-af071253236c · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 3

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.235836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d8510071-f806-4b6d-9121-4231c57102f6 · outbound

This paper cites Leave No Knowledge Behind During Knowledge Distillation: Towards Practical and Effective Knowledge Distillation for Code-Switching ASR Using Realistic Data.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Leave No Knowledge Behind During Knowledge Distillation: Towards Practical and Effective Knowledge Distillation for Code-Switching ASR Using Realistic Data

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.109502Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 849a1839-a2b3-4ab6-b4e8-b7b27250e8dd · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 5

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.223168Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 92a2c60d-67e5-4a02-8c95-3c14e67ca8a9 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.209412Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a8fb285a-384c-43dd-9b97-368b19340338 · outbound

This paper cites Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.126338Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7910d2f-1466-442e-b0be-e5f7507e0e3b · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 8

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.074441Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation aef16eda-0f4c-4622-8bab-1a88250caa90 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 9

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:33.020255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 17355a12-ad7f-43a1-8fda-b0da129b1e85 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 10

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.975102Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3280e51a-11d2-4c5e-9fff-ffd4755caa20 · outbound

This paper cites Pointer Sentinel Mixture Models.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Pointer Sentinel Mixture Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.142875Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9134e5f-dc58-4d76-9c23-4d9bf21d42eb · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 12

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.962514Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 5802684c-bfa0-4eb2-b5f8-e4d21dee2f05 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 13

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.949575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 41e6d0eb-05ef-4a0e-80b2-d84a7ad16097 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.935240Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 8e32829d-acb3-46ff-86be-c19eea4ca149 · outbound

This paper cites How Bad is Training on Synthetic Data? A Statistical Analysis of Language Model Collapse.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training How Bad is Training on Synthetic Data? A Statistical Analysis of Language Model Collapse

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.231826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e28b2583-8855-4b02-a74c-8d6f0e514a62 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.922611Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation bc96d4e4-5a4b-4402-8a2e-5e2dce5e1f27 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 17

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.909082Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d862d3ea-25f9-431b-afb8-59c192799891 · outbound

This paper cites Bias Amplification: Large Language Models as Increasingly Biased Media.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Bias Amplification: Large Language Models as Increasingly Biased Media

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.399625Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 191bf943-1063-4d3e-bcd5-9fde61496e9e · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.895607Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 74ab6cdd-16e9-4bb5-aeec-bd086bc18024 · outbound

This paper cites an unresolved cited work.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Unresolved cited work

Reference 20

Resolution: unresolved; raw_fallback, observed 2026-08-15T16:33:32.882940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation eeefc6f3-2139-4b9d-aba8-040d43b9760b · outbound

This paper cites Regurgitative Training: The Value of Real Data in Training Large Language Models.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Regurgitative Training: The Value of Real Data in Training Large Language Models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-15T16:33:32.412920Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7215eb4-c56d-4b5a-9f79-85bed0743941 · outbound

This paper cites Which term refers to enlightened beings in Buddhism? (A) Arhats (B) Bodhisattvas (C) Mahayana (D) TheravadaAnswer:.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training Which term refers to enlightened beings in Buddhism? (A) Arhats (B) Bodhisattvas (C) Mahayana (D) TheravadaAnswer:

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T16:33:32.869946Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 62f42bd0-0e4b-4c9a-bedf-6f8f615ea580 · outbound

This paper cites hand gestures.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training hand gestures

Reference 23

Resolution: malformed identifier; raw_fallback, observed 2026-08-15T16:33:32.826972Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 3dba84b8-cdf7-4171-91c4-cf2ecf677c74 · outbound

This paper cites The study focuses on automated analysis of language model behavior using computational methods and publicly available benchmarks, making IRB approval unnecessary.

Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training The study focuses on automated analysis of language model behavior using computational methods and publicly available benchmarks, making IRB approval unnecessary

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T16:33:32.738982Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Pith citing papers

Observation 5178baf1-7d3e-4dca-8b98-d5cbf95639fd · inbound

Self-Training Doesn't Flatten Language -- It Restructures It: Surface Markers Amplify While Deep Syntax Dies cites this paper.

Self-Training Doesn't Flatten Language -- It Restructures It: Surface Markers Amplify While Deep Syntax Dies Knowledge Collapse in LLMs: When Fluency Survives but Facts Fail under Recursive Synthetic Training

Reference 20

Resolution: verified exact; arxiv_id, observed 2026-05-21T05:49:41.096054Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.