Paper Citation Record · LEDGER

The Curse of Recursion: Training on Generated Data Makes Models Forget

As of 16 July 2026, this Paper Citation Record lists 12 of 12 outbound references and 36 inbound Pith citation observations for arXiv:2305.17493.

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

pith.paper-citation-record.v1
2305.17493 v3

Coverage vector

measured 12 of 12 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-11T11:50:26.030339Z

measured 48 of 48 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-07-15T06:30:58.975436+00:00

measured 36 of 36 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-14T09:09:29.470093Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-11T03:37:46.430605Z

Reference resolution

12 of 12 outbound references displayed

verified exact7
verified fuzzy3
unresolved0
parse uncertain0
malformed identifier0
metadata mismatch2

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 23d4a929-76d2-4c86-8e5d-28d9090d8464 · outbound

This paper cites Poisoning Attacks against Support Vector Machines.

The Curse of Recursion: Training on Generated Data Makes Models Forget Poisoning Attacks against Support Vector Machines

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:04:58.303410Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 5f53c9e5-c092-4503-8653-e9fd65f6000f · outbound

This paper cites Language models are few-shot learners.

The Curse of Recursion: Training on Generated Data Makes Models Forget Language models are few-shot learners

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:04:58.325033Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 5d5d74ee-b597-48e4-aa6b-82b284c6613d · outbound

This paper cites Poisoning and Backdooring Contrastive Learning.

The Curse of Recursion: Training on Generated Data Makes Models Forget Poisoning and Backdooring Contrastive Learning

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.294130Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 78c43c45-aa90-4656-8d54-93b6517f8980 · outbound

This paper cites Poisoning Web-Scale Training Datasets is Practical.

The Curse of Recursion: Training on Generated Data Makes Models Forget Poisoning Web-Scale Training Datasets is Practical

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.299671Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation a5aecff8-5655-47ad-83c9-ca2c265f3d2b · outbound

This paper cites BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding.

The Curse of Recursion: Training on Generated Data Makes Models Forget BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Reference 5

Resolution: metadata mismatch; local_arxiv, observed 2026-05-20T14:04:58.287573Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation f325026a-3e31-47f7-953b-677be17ae1b3 · outbound

This paper cites BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain.

The Curse of Recursion: Training on Generated Data Makes Models Forget BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain

Reference 6

Resolution: metadata mismatch; local_arxiv, observed 2026-05-20T14:04:58.318655Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation a4f2559e-b27b-47c8-862f-426c20e01408 · outbound

This paper cites RoBERTa: A Robustly Optimized BERT Pretraining Approach.

The Curse of Recursion: Training on Generated Data Makes Models Forget RoBERTa: A Robustly Optimized BERT Pretraining Approach

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:04:58.322139Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 2118cab2-e020-41e9-bca4-8a714e5daa46 · outbound

This paper cites Sponge examples: Energy-latency attacks on neural networks.

The Curse of Recursion: Training on Generated Data Makes Models Forget Sponge examples: Energy-latency attacks on neural networks

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:04:58.327624Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 95db7c8d-bbd5-401a-a15d-18a02058ea65 · outbound

This paper cites Energy and Policy Considerations for Deep Learning in NLP.

The Curse of Recursion: Training on Generated Data Makes Models Forget Energy and Policy Considerations for Deep Learning in NLP

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:04:58.308185Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 16f0e338-bc36-4fdc-a623-28f7f03f42c4 · outbound

This paper cites Three scenarios for continual learning.

The Curse of Recursion: Training on Generated Data Makes Models Forget Three scenarios for continual learning

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:04:58.311684Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 63bffabe-21b0-4685-860d-892509d8e277 · outbound

This paper cites OPT: Open Pre-trained Transformer Language Models.

The Curse of Recursion: Training on Generated Data Makes Models Forget OPT: Open Pre-trained Transformer Language Models

Reference 11

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:04:58.315231Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 68e8cd06-6b18-465a-8b5b-e0048992e0c9 · outbound

This paper cites The mean estimator and its standard deviation are calculated from running the procedure 10000 times.

The Curse of Recursion: Training on Generated Data Makes Models Forget The mean estimator and its standard deviation are calculated from running the procedure 10000 times

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-20T14:04:58.329959Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Pith citing papers

Observation 7cee5517-bc43-4446-872f-cc38eca1d699 · inbound

Textbooks Are All You Need cites this paper.

Textbooks Are All You Need The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 27

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 300bb500-be76-4e39-87ee-7f5b6e8a87c3 · inbound

Reinforced Self-Training (ReST) for Language Modeling cites this paper.

Reinforced Self-Training (ReST) for Language Modeling The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 22

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation b086f16b-ce87-4e45-923f-07a97fc78c5b · inbound

Evaluating Very Long-Term Conversational Memory of LLM Agents cites this paper.

Evaluating Very Long-Term Conversational Memory of LLM Agents The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 41

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 46a5c8b2-a400-42f2-b81c-7819b46e4343 · inbound

Scaling Synthetic Data Creation with 1,000,000,000 Personas cites this paper.

Scaling Synthetic Data Creation with 1,000,000,000 Personas The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation fb9d9938-43f4-45c1-96aa-4186b6aee03a · inbound

HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs cites this paper.

HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMs The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 66

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation cd3ce785-9a80-48c3-b774-60383cc2f983 · inbound

Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices cites this paper.

Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-05-23T01:05:16.448204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 32fb2bbe-94f6-4863-b5b5-6ad783894d4a · inbound

Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task cites this paper.

Your Brain on ChatGPT: Accumulation of Cognitive Debt when Using an AI Assistant for Essay Writing Task The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 116

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 84feba63-2c3a-4178-85c6-43e27581af93 · inbound

Losing our Tail, Again: (Un)Natural Selection & Multilingual LLMs cites this paper.

Losing our Tail, Again: (Un)Natural Selection & Multilingual LLMs The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 48

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 2e995268-57fa-456f-ad4c-4bdb7dd89bc5 · inbound

How to Model AI Agents as Personas?: Applying the Persona Ecosystem Playground to 41,300 Posts on Moltbook for Behavioral Insights cites this paper.

How to Model AI Agents as Personas?: Applying the Persona Ecosystem Playground to 41,300 Posts on Moltbook for Behavioral Insights The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-05-21T11:44:09.274914Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation fe180e36-da2c-4df8-a0df-e027ed63899a · inbound

Synthetic Data Generation for Training Diversified Commonsense Reasoning Models cites this paper.

Synthetic Data Generation for Training Diversified Commonsense Reasoning Models The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 3

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 16db3222-3c80-41e5-b323-0a983b7bd077 · inbound

Stabilizing Unsupervised Self-Evolution of MLLMs via Continuous Softened Retracing reSampling cites this paper.

Stabilizing Unsupervised Self-Evolution of MLLMs via Continuous Softened Retracing reSampling The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 2

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 54faf63d-32bb-46f6-ad9a-b597bc0f5357 · inbound

Content Platform GenAI Regulation via Compensation cites this paper.

Content Platform GenAI Regulation via Compensation The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 8447587a-6cce-4c23-8582-3172039612bd · inbound

RLSpoofer: A Lightweight Evaluator for LLM Watermark Spoofing Resilience cites this paper.

RLSpoofer: A Lightweight Evaluator for LLM Watermark Spoofing Resilience The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation f4e92bf8-0c0b-4781-8d80-97bc33a5c61b · inbound

Position: No Retroactive Cure for Infringement during Training cites this paper.

Position: No Retroactive Cure for Infringement during Training The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 9d9aaab9-1979-44b6-ba4e-2737d31363c6 · inbound

AgentSim: A Platform for Verifiable Agent-Trace Simulation cites this paper.

AgentSim: A Platform for Verifiable Agent-Trace Simulation The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 65bcb181-f9d1-492a-8521-a9074c9b46a7 · inbound

Perturbation Dose Responses in Recursive LLM Loops: Raw Switching, Stochastic Floors, and Persistent Escape under Append, Replace, and Dialog Updates cites this paper.

Perturbation Dose Responses in Recursive LLM Loops: Raw Switching, Stochastic Floors, and Persistent Escape under Append, Replace, and Dialog Updates The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 3

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 51cd5b87-5ea0-4b1d-8fad-bd5f99dda1a0 · inbound

Curated Synthetic Data Doesn't Have to Collapse: A Theoretical Study of Generative Retraining with Pluralistic Preferences cites this paper.

Curated Synthetic Data Doesn't Have to Collapse: A Theoretical Study of Generative Retraining with Pluralistic Preferences The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 120

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 7443ee2e-8b47-4b7b-8f89-3a299ecb3dfa · inbound

Annotations Mitigate Post-Training Mode Collapse cites this paper.

Annotations Mitigate Post-Training Mode Collapse The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 47

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation c350a236-db2c-4b5d-b0c0-fec446ed5b31 · inbound

EmbGen: Teaching with Reassembled Corpora cites this paper.

EmbGen: Teaching with Reassembled Corpora The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 76c496c5-2d84-4459-b712-86cab4f8d7eb · inbound

Base Models Look Human To AI Detectors cites this paper.

Base Models Look Human To AI Detectors The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 13

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 3bb198e8-e007-427e-8b63-6cc6518e77af · inbound

When Does Model Collapse Occur in Structured Interactive Learning? cites this paper.

When Does Model Collapse Occur in Structured Interactive Learning? The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 34

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T14:04:58.330652Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation a6ffde12-85d4-47ab-860b-ddce5cf48e47 · inbound

When Is Next-Token Prediction Useful? Marginalization, Ergodicity, Mixture Identifiability, Local Sufficiency, RAG, Tools, and Programming cites this paper.

When Is Next-Token Prediction Useful? Marginalization, Ergodicity, Mixture Identifiability, Local Sufficiency, RAG, Tools, and Programming The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-05-25T05:05:22.872955Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 6190e2fd-6fce-495a-b257-1f2d45ba3806 · inbound

When Is Next-Token Prediction Useful? Marginalization, Ergodicity, Mixture Identifiability, Local Sufficiency, RAG, Tools, and Programming cites this paper.

Reference 36

Resolution: verified exact; local_arxiv, observed 2026-06-30T16:35:12.678373Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 639b75cc-5c4c-46a6-a0ed-f2906bdc40d5 · inbound

AI-Associated Lexical Shifts Across 34 Languages: Cross-Lingual Convergence and Diachronic Uptake in News Writing cites this paper.

AI-Associated Lexical Shifts Across 34 Languages: Cross-Lingual Convergence and Diachronic Uptake in News Writing The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-06-29T23:04:02.183358Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation d63e45ff-0ba7-49a8-81b3-f007b7b6de9e · inbound

The Need for an External Observer Formalizing the Sufficiency Gap: A Mathematical Extension of Mixture Identifiability and Contextual Grounding in Sequence Models cites this paper.

The Need for an External Observer Formalizing the Sufficiency Gap: A Mathematical Extension of Mixture Identifiability and Contextual Grounding in Sequence Models The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 26

Resolution: verified exact; local_arxiv, observed 2026-06-29T18:13:49.161223Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 1b8d6c9a-75cf-42fe-9fee-47670e85d164 · inbound

Activation Steering for Synthetic Data Generation: The Role of Diversity in Downstream Safety Detection cites this paper.

Activation Steering for Synthetic Data Generation: The Role of Diversity in Downstream Safety Detection The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 46

Resolution: verified exact; local_arxiv, observed 2026-06-29T14:03:29.928867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 55a0a407-2bf8-488b-9748-26796fbf4651 · inbound

When AI Meets Wall Street: A Survey on Trustworthy AI in Fintech cites this paper.

When AI Meets Wall Street: A Survey on Trustworthy AI in Fintech The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 104

Resolution: verified exact; local_arxiv, observed 2026-06-29T14:43:31.578237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 77cf2212-78ef-4b5f-9bbe-1d4a64e49136 · inbound

Trust Region On-Policy Distillation cites this paper.

Trust Region On-Policy Distillation The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 74

Resolution: metadata mismatch; local_arxiv, observed 2026-07-01T20:56:13.475864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation bcdb89dc-ad31-413f-a2d3-2a1883144aac · inbound

Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolution cites this paper.

Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolution The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 39

Resolution: verified exact; local_arxiv, observed 2026-07-02T08:56:49.364784Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 97a86962-58c9-4862-b9ba-3efd88b028a5 · inbound

Beyond the Golden Teacher: Enhancing Graph Learning through LLM-GNN Co-teaching cites this paper.

Beyond the Golden Teacher: Enhancing Graph Learning through LLM-GNN Co-teaching The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 50

Resolution: verified exact; local_arxiv, observed 2026-07-03T09:07:47.684204Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 29079005-b695-40ad-a61d-18962f6663eb · inbound

When Sample Selection Bias Precipitates Model Collapse cites this paper.

When Sample Selection Bias Precipitates Model Collapse The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-07-03T23:59:06.088082Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 5317b82d-5567-43ba-9c81-bf87a19e3fb7 · inbound

DRIFT: Refining Instruction Data via On-Policy Data Attribution cites this paper.

DRIFT: Refining Instruction Data via On-Policy Data Attribution The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-07-03T19:18:54.735393Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 6de6acf8-3e85-41a4-ad20-0469cc012671 · inbound

Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning cites this paper.

Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 62

Resolution: metadata mismatch; local_arxiv, observed 2026-07-04T17:20:00.094509Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 031348a3-10fb-4d86-8796-860ca20e105a · inbound

The Crowded Embedding Space: A Mean-Field Mechanism for Emergent Marginalization in Retrieval-Augmented Agents cites this paper.

The Crowded Embedding Space: A Mean-Field Mechanism for Emergent Marginalization in Retrieval-Augmented Agents The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 14

Resolution: verified exact; local_arxiv, observed 2026-06-30T11:44:38.803910Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 05d9f0cc-9f43-4434-ba46-9fa1b8878f70 · inbound

Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking cites this paper.

Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 63

Resolution: verified exact; local_arxiv, observed 2026-07-11T03:37:46.466940Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-15T06:30:58.975436+00:00.

Observation 29b67a78-bf47-417d-8102-85d3dbb4ede8 · inbound

Weight-Adjusted Gradients Reveal Parameter Importance and Failure Modes in LLMs cites this paper.

Weight-Adjusted Gradients Reveal Parameter Importance and Failure Modes in LLMs The Curse of Recursion: Training on Generated Data Makes Models Forget

Reference 49

Resolution: unresolved; no resolver link, observed 2026-07-14T09:09:29.470093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.