mlsanityData to 14 Oct 2026

Do Androids Know They're Only Dreaming of Electric Sheep?

arXiv 2312.17249 ↗First posted 28 Dec 2023cs.CL cs.AI cs.LGcited by 67

On the map: Mechanistic interpretability → Hallucination detection from model internals

Abstract (arXiv)

We design probes trained on the internal representations of a transformer language model to predict its hallucinatory behavior on three grounded generation tasks. To train the probes, we annotate for span-level hallucination on both sampled (organic) and manually edited (synthetic) reference outputs. Our probes are narrowly trained and we find that they are sensitive to their training domain: they generalize poorly from one task to another or from synthetic to organic hallucinations. However, on in-domain data, they can reliably detect hallucinations at many transformer layers, achieving 95% of their peak performance as early as layer 4. Here, probing proves accurate for evaluating hallucination, outperforming several contemporary baselines and even surpassing an expert human annotator in response-level detection F1. Similarly, on span-level labeling, probes are on par or better than the expert annotator on two out of three generation tasks. Overall, we find that probing is a feasible and efficient alternative to language model hallucination evaluation when model states are available.

Key facts. Cited by 67 papers in the map (data through 14 Oct 2026). Placed in Hallucination detection from model internals, one of 13 directions in Mechanistic interpretability. Builds on 41 of its 56 references that are in the map. First posted on arXiv 28 Dec 2023.

Builds on

41 of 56 references.

References · 41

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2017Proximal Policy Optimization Algorithmscited by 15,215
  4. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  5. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  7. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  8. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  9. 2016Understanding intermediate layers using linear classifier probescited by 1,353
  10. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,336
  11. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  12. 2018Hierarchical Neural Story Generationcited by 1,193
29 more
  1. 2015Teaching Machines to Read and Comprehendcited by 1,049
  2. 2022Mass-Editing Memory in a Transformercited by 1,001
  3. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  4. 2019Evaluating the Factual Consistency of Abstractive Text Summarizationcited by 584
  5. 2021SummaC: Re-Visiting NLI-based Models for Inconsistency Detection in Summarizationcited by 407
  6. 2020Asking and Answering Questions to Evaluate the Factual Consistency of Summariescited by 378
  7. 2020FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarizationcited by 293
  8. 2017The E2E Dataset: New Challenges For End-to-End Generationcited by 253
  9. 2021QuestEval: Summarization Asks for Fact-based Evaluationcited by 244
  10. 2020Reducing conversational agents' overconfidence through linguistic calibrationcited by 210
  11. 2022Out-of-Distribution Detection and Selective Generation for Conditional Language Modelscited by 197
  12. 2019Neural Text Summarization: A Critical Evaluationcited by 189
  13. 2021Annotating and Modeling Fine-grained Factuality in Summarizationcited by 131
  14. 2020Evaluating Factuality in Generation with Dependency-level Entailmentcited by 122
  15. 2020GO FIGURE: A Meta Evaluation of Factuality in Summarizationcited by 76
  16. 2023WikiChat: Stopping the Hallucination of Large Language Model Chatbots by Few-Shot Grounding on Wikipediacited by 67
  17. 2023Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Modelscited by 67
  18. 2020Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Rewardcited by 66
  19. 2019Semantic Noise Matters for Neural Natural Language Generationcited by 64
  20. 2023Cognitive Dissonance: Why Do Language Model Outputs Disagree with Internal Representations of Truthfulness?cited by 62
  21. 2018UKP-Athene: Multi-Sentence Textual Entailment for Claim Verificationcited by 60
  22. 2023Understanding and Detecting Hallucinations in Neural Machine Translation via Model Introspectioncited by 44
  23. 2020Towards Ecologically Valid Research on Language User Interfacescited by 42
  24. 2021Rome was built in 1776: A Case Study on Factual Correctness in Knowledge-Grounded Response Generationcited by 29
  25. 2023A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepediacited by 24
  26. 2023LongDocFACTScore: Evaluating the Factuality of Long Document Abstractive Summarisationcited by 16
  27. 2022BUMP: A Benchmark of Unfaithful Minimal Pairs for Meta-Evaluation of Faithfulness Metricscited by 7
  28. 2022Exact Paired-Permutation Testing for Structured Test Statisticscited by 6
  29. 2023Diversity Measures: Domain-Independent Proxies for Failure in Language Model Queriescited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 17; 2025: 28; 2026: 22.

67 papers.

Sort:

By area:

  1. 2024Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaborationcited by 180
  2. 2024Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Modelscited by 115
  3. 2024A Primer on the Inner Workings of Transformer-based Language Modelscited by 85
  4. 2024Uncertainty Estimation and Quantification for LLMs: A Simple Supervised Approachcited by 76
  5. 2024Get my drift? Catching LLM Task Drift with Activation Deltascited by 72
  6. 2024Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluationcited by 57
  7. 2024Insights into LLM Long-Context Failures: When Transformers Know but Don't Tellcited by 36
  8. 2025Real-Time Detection of Hallucinated Entities in Long-Form Generationcited by 34
  9. 2025ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMscited by 32
  10. 2025Trust Me, I'm Wrong: LLMs Hallucinate with Certainty Despite Knowing the Answercited by 27
  11. 2024On the Universal Truthfulness Hyperplane Inside LLMscited by 24
  12. 2024Towards Uncovering How Large Language Model Works: An Explainability Perspectivecited by 22
38 more
  1. 2025Representation Engineering for Large-Language Models: Survey and Research Challengescited by 21
  2. 2025A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads for Hallucination Detection in LLM Outputscited by 21
  3. 2024Constructing Benchmarks and Interventions for Combating Hallucinations in LLMscited by 19
  4. 2025Hallucination Detection in LLMs with Topological Divergence on Attention Graphscited by 19
  5. 2025Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of Large Language Modelscited by 18
  6. 2025Probing the Geometry of Truth: Consistency and Generalization of Truth Directions in LLMs Across Logical Transformations and Question Answering Taskscited by 18
  7. 2025Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Headscited by 16
  8. 2024Distinguishing Ignorance from Error in LLM Hallucinationscited by 14
  9. 2025When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQAcited by 14
  10. 2026HALP: Detecting Hallucinations in Vision-Language Models without Generating a Single Tokencited by 14
  11. 2024REAL Sampling: Boosting Factuality and Diversity of Open-Ended Generation via Asymptotic Entropycited by 11
  12. 2024FactCheckmate: Preemptively Detecting and Mitigating Hallucinations in LMscited by 11
  13. 2025Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safetycited by 10
  14. 2024Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Modelscited by 9
  15. 2025Always Tell Me The Odds: Fine-grained Conditional Probability Estimationcited by 9
  16. 2025The Geometries of Truth Are Orthogonal Across Taskscited by 9
  17. 2025LLM Knowledge is Brittle: Truthfulness Representations Rely on Superficial Resemblancecited by 9
  18. 2024Prompt-Guided Internal States for Hallucination Detection of Large Language Modelscited by 8
  19. 2026The Truthfulness Spectrum Hypothesiscited by 8
  20. 2025When Do LLMs Admit Their Mistakes? Understanding The Role Of Model Belief In Retractioncited by 6
  21. 2025HACK: Hallucinations Along Certainty and Knowledge Axescited by 6
  22. 2024Shared Imagination: LLMs Hallucinate Alikecited by 5
  23. 2026Hallucinations Undermine Trust; Metacognition is a Way Forwardcited by 5
  24. 2025RePPL: Recalibrating Perplexity by Uncertainty in Semantic Propagation and Language Generation for Explainable QA Hallucination Detectioncited by 4
  25. 2025The Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factualitycited by 4
  26. 2025Cross-Layer Attention Probing for Fine-Grained Hallucination Detectioncited by 4
  27. 2025Localized Calibrated Uncertainty in Code Language Modelscited by 3
  28. 2025Data-efficient Meta-models for Evaluation of Context-based Questions and Answers in LLMscited by 2
  29. 2025Shapley Uncertainty in Natural Language Generationcited by 2
  30. 2026HalluZig: Hallucination Detection using Zigzag Persistencecited by 2
  31. 2026INTRYGUE: Induction-Aware Entropy Gating for Reliable RAG Uncertainty Estimationcited by 2
  32. 2026Process Supervision of Confidence Margin for Calibrated LLM Reasoningcited by 2
  33. 2026From Signals to Transfer: A Factorised Study of Probe-Based Uncertainty Estimation in Large Language Modelscited by 2
  34. 2026HalluTruthQA: A Fine-Grained Benchmark for Hallucination Detection, Localization, and Explanation in Arabic Question Answeringcited by 2
  35. 2026Language Models Encode the Contextual Truth of Propositionscited by 2
  36. 2024Listening to the Wise Few: Query-Key Alignment Unlocks Latent Correct Answers in Large Language Modelscited by 1
  37. 2025Exploring the generalization of LLM truth directions on conversational formatscited by 1
  38. 2025ALIEN: Aligned Entropy Head for Improving Uncertainty Estimation of LLMscited by 1

50 most cited of 67 All 67 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Detection Without Correction: A Robust Asymmetry in Activation-Based Hallucination Probing
  2. 2025A Single Direction of Truth: An Observer Model's Linear Residual Probe Exposes and Steers Contextual Hallucinations
  3. 2026External Observers May See More Clearly: Cross-Model Span-Level Hallucination Detection in Large Language Models via Hidden State Probing
  4. 2026VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck
  5. 2025MTRE: Multi-Token Reliability Estimation for Hallucination Detection in VLMs
  6. 2026HalluScope: Fine-grained Hallucination Diagnosis for Multimodal Large Language Models
  7. 2026HalluSAE: Detecting Hallucinations in Large Language Models via Sparse Auto-Encoders
  8. 2026Grounding Probes: Generator-Independent Hallucination Detection from Observer Model Hidden States