mlsanityData to 14 Oct 2026

Semantic Noise Matters for Neural Natural Language Generation

arXiv 1911.03905 ↗First posted 10 Nov 2019cs.CLcited by 64

Not on the map. Why?

Abstract (arXiv)

Neural natural language generation (NNLG) systems are known for their pathological outputs, i.e. generating text which is unrelated to the input specification. In this paper, we show the impact of semantic noise on state-of-the-art NNLG models which implement different semantic control mechanisms. We find that cleaned data can improve semantic correctness by up to 97%, while maintaining fluency. We also find that the most common error is omitting information, rather than hallucination.

Key facts. Cited by 64 papers in the map (data through 14 Oct 2026). First posted on arXiv 10 Nov 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 15; 2022: 18; 2023: 17; 2024: 8; 2025: 3; 2026: 3.

64 papers.

Sort:

By area:

  1. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  2. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  3. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
  4. 2022GLM-130B: An Open Bilingual Pre-trained Modelcited by 843
  5. 2022UL2: Unifying Language Learning Paradigmscited by 295
  6. 2021The GEM Benchmark: Natural Language Generation, its Evaluation and Metricscited by 239
  7. 2021SPoT: Better Frozen Model Adaptation through Soft Prompt Transfercited by 235
  8. 2021ExT5: Towards Extreme Multi-Task Scaling for Transfer Learningcited by 161
  9. 2022Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Textcited by 155
  10. 2021CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLPcited by 142
  11. 2023Unified Demonstration Retriever for In-Context Learningcited by 142
  12. 2022Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacycited by 100
38 more
  1. 2024Representation Noising: A Defence Mechanism Against Harmful Finetuningcited by 96
  2. 2023UPRISE: Universal Prompt Retrieval for Improving Zero-Shot Evaluationcited by 75
  3. 2024In-context Learning with Retrieved Demonstrations for Language Models: A Surveycited by 70
  4. 2023Do Androids Know They're Only Dreaming of Electric Sheep?cited by 67
  5. 2023A Multi-Task Embedder For Retrieval Augmented LLMscited by 64
  6. 2023Learning to Retrieve In-Context Examples for Large Language Modelscited by 60
  7. 2022AlexaTM 20B: Few-Shot Learning Using a Large-Scale Multilingual Seq2Seq Modelcited by 57
  8. 2023Adapting Large Language Models to Domains via Reading Comprehensioncited by 50
  9. 2023Decoder-Only or Encoder-Decoder? Interpreting Language Model as a Regularized Encoder-Decodercited by 44
  10. 2021Controlling Hallucinations at Word Level in Data-to-Text Generationcited by 41
  11. 2022GEMv2: Multilingual NLG Benchmarking in a Single Line of Codecited by 36
  12. 2024AI Hallucinations: A Misnomer Worth Clarifyingcited by 34
  13. 2022Learning to Revise References for Faithful Summarizationcited by 28
  14. 2021Exploring Universal Intrinsic Task Subspace via Prompt Tuningcited by 25
  15. 2021Structural Information Preserving for Graph-to-Text Generationcited by 24
  16. 2021Control Prefixes for Parameter-Efficient Text Generationcited by 24
  17. 2021Underreporting of errors in NLG output, and what to do about itcited by 19
  18. 2022Neural Pipeline for Zero-Shot Data-to-Text Generationcited by 19
  19. 2021Automatic Construction of Evaluation Suites for Natural Language Generation Datasetscited by 18
  20. 2023Small Language Models Improve Giants by Rewriting Their Outputscited by 18
  21. 2024Beyond Traditional Benchmarks: Analyzing Behaviors of Open LLMs on Data-to-Text Generationcited by 16
  22. 2022Eliciting and Understanding Cross-Task Skills with Task-Level Mixture-of-Expertscited by 12
  23. 2022Contrastive Error Attribution for Finetuned Language Modelscited by 12
  24. 2023Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning?cited by 11
  25. 2022Innovations in Neural Data-to-text Generation: A Surveycited by 10
  26. 2022TaTa: A Multilingual Table-to-Text Dataset for African Languagescited by 10
  27. 2024Detectors for Safe and Reliable LLMs: Implementations, Uses, and Limitationscited by 10
  28. 2021Search and Learn: Improving Semantic Coverage for Data-to-Text Generationcited by 8
  29. 2022What Makes Data-to-Text Generation Hard for Pretrained Language Models?cited by 8
  30. 2022Self-supervised Graph Masking Pre-training for Graph-to-Text Generationcited by 8
  31. 2024Se2: Sequential Example Selection for In-Context Learningcited by 8
  32. 2021AGGGEN: Ordering and Aggregating while Generatingcited by 7
  33. 2023TabGenie: A Toolkit for Table-to-Text Generationcited by 7
  34. 2021BEAMetrics: A Benchmark for Language Generation Evaluation Evaluationcited by 6
  35. 2022Mind the Labels: Describing Relations in Knowledge Graphs With Pretrained Modelscited by 6
  36. 2025Dynamic Task Vector Grouping for Efficient Multi-Task Prompt Tuningcited by 6
  37. 2023Sequence-Level Certainty Reduces Hallucination In Knowledge-Grounded Dialogue Generationcited by 5
  38. 2023Beyond Output Matching: Bidirectional Alignment for Enhanced In-Context Learningcited by 5

50 most cited of 64 All 64 (JSON)