# Reference papers

127 papers on uncertainty estimation, misclassification / failure detection, OOD detection and internal-state probing, 2020–2026 (a few older iconic ones). PDFs are not committed; `bash reference_papers/fetch.sh` re-downloads them from `manifest.tsv`. Layout: `reference_papers/<venue>/<year>_<slug>.pdf`; `Findings`, demo and workshop papers live under their parent conference with a `findings_`/`demo_`/`blackboxnlp_` prefix. LaTeX templates for the target venues are under `templates/`.

## NAACL (19)

| year | file | title | source |
|---|---|---|---|
| 2021 | `2021_noisy_ner_confidence.pdf` | Noisy-Labeled NER with Confidence Estimation | [acl:2021.naacl-main.269](https://aclanthology.org/2021.naacl-main.269/) |
| 2022 | `2022_findings_pcee_bert_early_exit.pdf` | PCEE-BERT: Accelerating BERT Inference via Patient and Confident Early Exiting | [acl:2022.findings-naacl.25](https://aclanthology.org/2022.findings-naacl.25/) |
| 2022 | `2022_wu_revisit_overconfidence_ood.pdf` | Revisit Overconfidence for OOD Detection: Reassigned Contrastive Learning with Adaptive Class-dependent Threshold | [acl:2022.naacl-main.307](https://aclanthology.org/2022.naacl-main.307/) |
| 2024 | `2024_findings_he_ue_sequential_labeling.pdf` | Uncertainty Estimation on Sequential Labeling via Uncertainty Transmission | [acl:2024.findings-naacl.180](https://aclanthology.org/2024.findings-naacl.180/) |
| 2024 | `2024_geng_survey_confidence_calibration_llm.pdf` | A Survey of Confidence Estimation and Calibration in Large Language Models | [acl:2024.naacl-long.366](https://aclanthology.org/2024.naacl-long.366/) |
| 2024 | `2024_kabra_program_aided_know.pdf` | Program-Aided Reasoners (Better) Know What They Know | [acl:2024.naacl-long.125](https://aclanthology.org/2024.naacl-long.125/) |
| 2024 | `2024_ling_uq_icl.pdf` | Uncertainty Quantification for In-Context Learning of Large Language Models | [acl:2024.naacl-long.184](https://aclanthology.org/2024.naacl-long.184/) |
| 2024 | `2024_zhang_calibration_icl.pdf` | A Study on the Calibration of In-context Learning | [acl:2024.naacl-long.340](https://aclanthology.org/2024.naacl-long.340/) |
| 2025 | `2025_belief_tree_hallucination.pdf` | A Probabilistic Framework for LLM Hallucination Detection via Belief Tree Propagation | [acl:2025.naacl-long.158](https://aclanthology.org/2025.naacl-long.158/) |
| 2025 | `2025_beyond_logit_lens_hallucination.pdf` | Beyond Logit Lens: Contextual Embeddings for Robust Hallucination Detection & Grounding in VLMs | [acl:2025.naacl-long.488](https://aclanthology.org/2025.naacl-long.488/) |
| 2025 | `2025_findings_llm_ood_survey.pdf` | Large Language Models for Anomaly and Out-of-Distribution Detection: A Survey | [acl:2025.findings-naacl.333](https://aclanthology.org/2025.findings-naacl.333/) |
| 2025 | `2025_findings_long_context_hallucination.pdf` | Towards Long Context Hallucination Detection | [acl:2025.findings-naacl.436](https://aclanthology.org/2025.findings-naacl.436/) |
| 2025 | `2025_findings_maqa_data_uncertainty.pdf` | MAQA: Evaluating Uncertainty Quantification in LLMs Regarding Data Uncertainty | [acl:2025.findings-naacl.325](https://aclanthology.org/2025.findings-naacl.325/) |
| 2025 | `2025_findings_sycophancy_ue.pdf` | Accounting for Sycophancy in Language Model Uncertainty Estimation | [acl:2025.findings-naacl.438](https://aclanthology.org/2025.findings-naacl.438/) |
| 2025 | `2025_findings_trainable_scoring_ue.pdf` | Do Not Design, Learn: A Trainable Scoring Function for Uncertainty Estimation in Generative LLMs | [acl:2025.findings-naacl.41](https://aclanthology.org/2025.findings-naacl.41/) |
| 2025 | `2025_findings_vazhentsev_knn_ue.pdf` | Efficient Nearest Neighbor based Uncertainty Estimation for NLP Tasks | [acl:2025.findings-naacl.246](https://aclanthology.org/2025.findings-naacl.246/) |
| 2025 | `2025_mice_model_internal_confidence.pdf` | MICE for CATs: Model-Internal Confidence Estimation for Calibrating Agents with Tools | [acl:2025.naacl-long.615](https://aclanthology.org/2025.naacl-long.615/) |
| 2025 | `2025_sciurus_uncertainty_circuits.pdf` | SCIURus: Shared Circuits for Interpretable Uncertainty Representations in Language Models | [acl:2025.naacl-long.618](https://aclanthology.org/2025.naacl-long.618/) |
| 2025 | `2025_vazhentsev_token_density_uq.pdf` | Token-Level Density-Based Uncertainty Quantification Methods for Eliciting Truthfulness of LLMs | [acl:2025.naacl-long.113](https://aclanthology.org/2025.naacl-long.113/) |

## ACL (30)

| year | file | title | source |
|---|---|---|---|
| 2020 | `2020_hendrycks_pretrained_ood_robustness.pdf` | Pretrained Transformers Improve Out-of-Distribution Robustness | [acl:2020.acl-main.244](https://aclanthology.org/2020.acl-main.244/) |
| 2020 | `2020_kamath_selective_qa.pdf` | Selective Question Answering under Domain Shift | [acl:2020.acl-main.503](https://aclanthology.org/2020.acl-main.503/) |
| 2021 | `2021_xin_art_of_abstention.pdf` | The Art of Abstention: Selective Prediction and Error Regularization for NLP | [acl:2021.acl-long.84](https://aclanthology.org/2021.acl-long.84/) |
| 2021 | `2021_xu_unsupervised_ood_pretrained.pdf` | Unsupervised Out-of-Domain Detection via Pre-trained Transformers | [acl:2021.acl-long.85](https://aclanthology.org/2021.acl-long.85/) |
| 2022 | `2022_findings_varshney_selective_prediction.pdf` | Investigating Selective Prediction Approaches Across Several Tasks in IID, OOD, and Adversarial Settings | [acl:2022.findings-acl.158](https://aclanthology.org/2022.findings-acl.158/) |
| 2022 | `2022_park_calibration_mixup_aum.pdf` | On the Calibration of Pre-trained Language Models using Mixup Guided by Area Under the Margin | [acl:2022.acl-long.368](https://aclanthology.org/2022.acl-long.368/) |
| 2022 | `2022_vazhentsev_ue_misclassification.pdf` | Uncertainty Estimation of Transformer Predictions for Misclassification Detection | [acl:2022.acl-long.566](https://aclanthology.org/2022.acl-long.566/) |
| 2022 | `2022_zhou_knn_contrastive_ood.pdf` | KNN-Contrastive Learning for Out-of-Domain Intent Classification | [acl:2022.acl-long.352](https://aclanthology.org/2022.acl-long.352/) |
| 2023 | `2023_chen_close_look_calibration_plm.pdf` | A Close Look into the Calibration of Pre-trained Language Models | [acl:2023.acl-long.75](https://aclanthology.org/2023.acl-long.75/) |
| 2023 | `2023_chen_evaluation_selective_prediction.pdf` | On the Evaluation of Neural Selective Prediction Methods for NLP | [acl:2023.acl-long.437](https://aclanthology.org/2023.acl-long.437/) |
| 2023 | `2023_findings_pseudo_outlier_exposure.pdf` | Pseudo Outlier Exposure for Out-of-Distribution Detection using Pretrained Transformers | [acl:2023.findings-acl.95](https://aclanthology.org/2023.findings-acl.95/) |
| 2023 | `2023_findings_yin_llms_know_dont_know.pdf` | Do Large Language Models Know What They Don't Know? | [acl:2023.findings-acl.551](https://aclanthology.org/2023.findings-acl.551/) |
| 2023 | `2023_uppaal_is_finetuning_needed.pdf` | Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection | [acl:2023.acl-long.717](https://aclanthology.org/2023.acl-long.717/) |
| 2023 | `2023_vazhentsev_hybrid_uq.pdf` | Hybrid Uncertainty Quantification for Selective Text Classification in Ambiguous Tasks | [acl:2023.acl-long.652](https://aclanthology.org/2023.acl-long.652/) |
| 2024 | `2024_duan_shifting_attention_relevance.pdf` | Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form LLMs | [acl:2024.acl-long.276](https://aclanthology.org/2024.acl-long.276/) |
| 2024 | `2024_findings_fadeeva_token_uq_factcheck.pdf` | Fact-Checking the Output of LLMs via Token-Level Uncertainty Quantification | [acl:2024.findings-acl.558](https://aclanthology.org/2024.findings-acl.558/) |
| 2024 | `2024_findings_su_mind_internal_states.pdf` | Unsupervised Real-Time Hallucination Detection based on the Internal States of LLMs | [acl:2024.findings-acl.854](https://aclanthology.org/2024.findings-acl.854/) |
| 2024 | `2024_mars_meaning_aware_scoring.pdf` | MARS: Meaning-Aware Response Scoring for Uncertainty Estimation in Generative LLMs | [acl:2024.acl-long.419](https://aclanthology.org/2024.acl-long.419/) |
| 2025 | `2025_findings_survey_ue_llm.pdf` | A Survey of Uncertainty Estimation Methods on Large Language Models | [acl:2025.findings-acl.1101](https://aclanthology.org/2025.findings-acl.1101/) |
| 2025 | `2025_hidden_states_hiding_something.pdf` | Are the Hidden States Hiding Something? Testing the Limits of Factuality-Encoding Capabilities in LLMs | [acl:2025.acl-long.304](https://aclanthology.org/2025.acl-long.304/) |
| 2025 | `2025_icr_probe_hidden_state_dynamics.pdf` | ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs | [acl:2025.acl-long.880](https://aclanthology.org/2025.acl-long.880/) |
| 2025 | `2025_internal_states_knowledge_boundary.pdf` | Towards Fully Exploiting LLM Internal States to Enhance Knowledge Boundary Perception | [acl:2025.acl-long.1184](https://aclanthology.org/2025.acl-long.1184/) |
| 2025 | `2025_overconfident_and_other_lies.pdf` | Your Model is Overconfident, and Other Lies We Tell Ourselves | [acl:2025.acl-long.269](https://aclanthology.org/2025.acl-long.269/) |
| 2025 | `2025_prompt_guided_internal_states.pdf` | Prompt-Guided Internal States for Hallucination Detection of Large Language Models | [acl:2025.acl-long.1058](https://aclanthology.org/2025.acl-long.1058/) |
| 2025 | `2025_short_uq_evaluation_spurious.pdf` | Revisiting Uncertainty Quantification Evaluation in Language Models: Spurious Interactions with Response Length | [acl:2025.acl-short.60](https://aclanthology.org/2025.acl-short.60/) |
| 2025 | `2025_zhang_llm_ue_in_the_wild.pdf` | Reconsidering LLM Uncertainty Estimation Methods in the Wild | [acl:2025.acl-long.1429](https://aclanthology.org/2025.acl-long.1429/) |
| 2026 | `2026_findings_confidence_evaluation_pitfalls.pdf` | Evaluation Pitfalls and Sparsity Limitations in LLM-based Confidence Estimates for Classification | [acl:2026.findings-acl.1671](https://aclanthology.org/2026.findings-acl.1671/) |
| 2026 | `2026_findings_internal_states_recall.pdf` | Do LLMs Really Know What They Don't Know? Internal States Mainly Reflect Knowledge Recall | [acl:2026.findings-acl.34](https://aclanthology.org/2026.findings-acl.34/) |
| 2026 | `2026_reprobe_internal_states.pdf` | ReProbe: Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States | [acl:2026.acl-long.536](https://aclanthology.org/2026.acl-long.536/) |
| 2026 | `2026_topological_divergence_attention_hallucination.pdf` | Hallucination Detection in LLMs with Topological Divergence on Attention Graphs | [acl:2026.acl-long.704](https://aclanthology.org/2026.acl-long.704/) |

## EMNLP (25)

| year | file | title | source |
|---|---|---|---|
| 2020 | `2020_desai_calibration_pretrained_transformers.pdf` | Calibration of Pre-trained Transformers | [acl:2020.emnlp-main.21](https://aclanthology.org/2020.emnlp-main.21/) |
| 2020 | `2020_he_accurate_ue_text_classification.pdf` | Towards More Accurate Uncertainty Estimation In Text Classification | [acl:2020.emnlp-main.671](https://aclanthology.org/2020.emnlp-main.671/) |
| 2020 | `2020_kong_calibrated_finetuning.pdf` | Calibrated Language Model Fine-Tuning for In- and Out-of-Distribution Data | [acl:2020.emnlp-main.102](https://aclanthology.org/2020.emnlp-main.102/) |
| 2021 | `2021_arora_types_of_ood_texts.pdf` | Types of Out-of-Distribution Texts and How to Detect Them | [acl:2021.emnlp-main.835](https://aclanthology.org/2021.emnlp-main.835/) |
| 2021 | `2021_kushnareva_topology_attention.pdf` | Artificial Text Detection via Examining the Topology of Attention Maps | [acl:2021.emnlp-main.50](https://aclanthology.org/2021.emnlp-main.50/) |
| 2021 | `2021_zhou_contrastive_ood.pdf` | Contrastive Out-of-Distribution Detection for Pretrained Transformers | [acl:2021.emnlp-main.84](https://aclanthology.org/2021.emnlp-main.84/) |
| 2022 | `2022_findings_cherniavskii_acceptability_topology.pdf` | Acceptability Judgements via Examining the Topology of Attention Maps | [acl:2022.findings-emnlp.7](https://aclanthology.org/2022.findings-emnlp.7/) |
| 2022 | `2022_findings_cho_implicit_layer_ensemble_ood.pdf` | Enhancing OOD Detection in NLU via Implicit Layer Ensemble | [acl:2022.findings-emnlp.55](https://aclanthology.org/2022.findings-emnlp.55/) |
| 2022 | `2022_findings_holistic_sentence_embeddings_ood.pdf` | Holistic Sentence Embeddings for Better Out-of-Distribution Detection | [acl:2022.findings-emnlp.497](https://aclanthology.org/2022.findings-emnlp.497/) |
| 2022 | `2022_findings_ulmer_uncertainty_calibration_nlp.pdf` | Exploring Predictive Uncertainty and Calibration in NLP | [acl:2022.findings-emnlp.198](https://aclanthology.org/2022.findings-emnlp.198/) |
| 2022 | `2022_findings_xiao_uq_plm_empirical.pdf` | Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysis | [acl:2022.findings-emnlp.538](https://aclanthology.org/2022.findings-emnlp.538/) |
| 2023 | `2023_demo_lm_polygraph.pdf` | LM-Polygraph: Uncertainty Estimation for Language Models | [acl:2023.emnlp-demo.41](https://aclanthology.org/2023.emnlp-demo.41/) |
| 2023 | `2023_findings_azaria_internal_state_lying.pdf` | The Internal State of an LLM Knows When It's Lying | [acl:2023.findings-emnlp.68](https://aclanthology.org/2023.findings-emnlp.68/) |
| 2023 | `2023_findings_document_ood_critical.pdf` | A Critical Analysis of Document Out-of-Distribution Detection | [acl:2023.findings-emnlp.332](https://aclanthology.org/2023.findings-emnlp.332/) |
| 2023 | `2023_flats_likelihood_ratio_ood.pdf` | FLatS: Principled Out-of-Distribution Detection with Feature-Based Likelihood Ratio Score | [acl:2023.emnlp-main.554](https://aclanthology.org/2023.emnlp-main.554/) |
| 2023 | `2023_slobodkin_unanswerability_hidden_states.pdf` | The Curious Case of Hallucinatory (Un)answerability: Finding Truths in the Hidden States of LLMs | [acl:2023.emnlp-main.220](https://aclanthology.org/2023.emnlp-main.220/) |
| 2023 | `2023_tian_just_ask_for_calibration.pdf` | Just Ask for Calibration | [acl:2023.emnlp-main.330](https://aclanthology.org/2023.emnlp-main.330/) |
| 2024 | `2024_blackboxnlp_ji_internal_states_risk.pdf` | LLM Internal States Reveal Hallucination Risk Faced With a Query | [acl:2024.blackboxnlp-1.6](https://aclanthology.org/2024.blackboxnlp-1.6/) |
| 2024 | `2024_embedding_gradient_say_wrong.pdf` | Embedding and Gradient Say Wrong: A White-Box Method for Hallucination Detection | [acl:2024.emnlp-main.116](https://aclanthology.org/2024.emnlp-main.116/) |
| 2024 | `2024_findings_internal_inspector.pdf` | InternalInspector: Robust Confidence Estimation in LLMs through Internal States | [acl:2024.findings-emnlp.751](https://aclanthology.org/2024.findings-emnlp.751/) |
| 2024 | `2024_gottesman_geva_knowledge_without_token.pdf` | Estimating Knowledge in LLMs Without Generating a Single Token | [acl:2024.emnlp-main.232](https://aclanthology.org/2024.emnlp-main.232/) |
| 2025 | `2025_findings_representation_detectors_fail_ood.pdf` | Representation-based Broad Hallucination Detectors Fail to Generalize Out of Distribution | [acl:2025.findings-emnlp.952](https://aclanthology.org/2025.findings-emnlp.952/) |
| 2025 | `2025_illusion_of_progress_hallucination.pdf` | The Illusion of Progress: Re-evaluating Hallucination Detection in LLMs | [acl:2025.emnlp-main.1761](https://aclanthology.org/2025.emnlp-main.1761/) |
| 2025 | `2025_kim_layerwise_information_deficiency.pdf` | Detecting LLM Hallucination Through Layer-wise Information Deficiency | [acl:2025.emnlp-main.1644](https://aclanthology.org/2025.emnlp-main.1644/) |
| 2025 | `2025_uq_heads_predict_question.pdf` | A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads | [acl:2025.emnlp-main.1809](https://aclanthology.org/2025.emnlp-main.1809/) |

## TACL (1)

| year | file | title | source |
|---|---|---|---|
| 2025 | `2025_vashurin_lm_polygraph_benchmark.pdf` | Benchmarking Uncertainty Quantification Methods for LLMs with LM-Polygraph | [acl:2025.tacl-1.11](https://aclanthology.org/2025.tacl-1.11/) |

## EACL (2)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_spuq_perturbation_uq.pdf` | SPUQ: Perturbation-Based Uncertainty Quantification for Large Language Models | [acl:2024.eacl-long.143](https://aclanthology.org/2024.eacl-long.143/) |
| 2026 | `2026_confidence_estimators_reasoning_models.pdf` | How Reliable are Confidence Estimators for Large Reasoning Models? | [acl:2026.eacl-long.78](https://aclanthology.org/2026.eacl-long.78/) |

## COLING (1)

| year | file | title | source |
|---|---|---|---|
| 2020 | `2020_mahalanobis_generative_ood.pdf` | A Deep Generative Distance-Based Classifier for Out-of-Domain Detection with Mahalanobis Space | [acl:2020.coling-main.125](https://aclanthology.org/2020.coling-main.125/) |

## ICLR (12)

| year | file | title | source |
|---|---|---|---|
| 2021 | `2021_sehwag_ssd.pdf` | SSD: A Unified Framework for Self-Supervised Outlier Detection | [arxiv:2103.12051](https://arxiv.org/abs/2103.12051) |
| 2023 | `2023_burns_discovering_latent_knowledge.pdf` | Discovering Latent Knowledge in Language Models Without Supervision | [arxiv:2212.03827](https://arxiv.org/abs/2212.03827) |
| 2023 | `2023_galil_selective_prediction_imagenet.pdf` | What Can We Learn From The Selective Prediction And Uncertainty Estimation Performance Of 523 Imagenet Classifiers | [arxiv:2302.11874](https://arxiv.org/abs/2302.11874) |
| 2023 | `2023_he_preserving_pretrained_features_calibration.pdf` | Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models | [arxiv:2305.19249](https://arxiv.org/abs/2305.19249) |
| 2023 | `2023_jaeger_failure_detection_evaluation.pdf` | A Call to Reflect on Evaluation Practices for Failure Detection in Image Classification | [arxiv:2211.15259](https://arxiv.org/abs/2211.15259) |
| 2023 | `2023_kuhn_semantic_uncertainty.pdf` | Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in NLG | [arxiv:2302.09664](https://arxiv.org/abs/2302.09664) |
| 2023 | `2023_ming_hyperspherical_ood.pdf` | How to Exploit Hyperspherical Embeddings for Out-of-Distribution Detection? | [arxiv:2203.04450](https://arxiv.org/abs/2203.04450) |
| 2023 | `2023_ren_rmd_selective_generation.pdf` | Out-of-Distribution Detection and Selective Generation for Conditional Language Models | [arxiv:2209.15558](https://arxiv.org/abs/2209.15558) |
| 2024 | `2024_chen_inside.pdf` | INSIDE: LLMs' Internal States Retain the Power of Hallucination Detection | [arxiv:2402.03744](https://arxiv.org/abs/2402.03744) |
| 2024 | `2024_chuang_dola.pdf` | DoLa: Decoding by Contrasting Layers Improves Factuality in LLMs | [arxiv:2309.03883](https://arxiv.org/abs/2309.03883) |
| 2024 | `2024_xiong_llms_express_uncertainty.pdf` | Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation | [arxiv:2306.13063](https://arxiv.org/abs/2306.13063) |
| 2025 | `2025_orgad_llms_know_more.pdf` | LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations | [arxiv:2410.02707](https://arxiv.org/abs/2410.02707) |

## NeurIPS (7)

| year | file | title | source |
|---|---|---|---|
| 2020 | `2020_liu_energy_ood.pdf` | Energy-based Out-of-distribution Detection | [arxiv:2010.03759](https://arxiv.org/abs/2010.03759) |
| 2020 | `2020_liu_sngp.pdf` | Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness | [arxiv:2006.10108](https://arxiv.org/abs/2006.10108) |
| 2021 | `2021_fort_exploring_limits_ood.pdf` | Exploring the Limits of Out-of-Distribution Detection | [arxiv:2106.03004](https://arxiv.org/abs/2106.03004) |
| 2022 | `2022_kotelevskii_nuq.pdf` | Nonparametric Uncertainty Quantification for Single Deterministic Neural Network | [arxiv:2202.03101](https://arxiv.org/abs/2202.03101) |
| 2023 | `2023_li_inference_time_intervention.pdf` | Inference-Time Intervention: Eliciting Truthful Answers from a Language Model | [arxiv:2306.03341](https://arxiv.org/abs/2306.03341) |
| 2023 | `2023_tulchinskii_intrinsic_dimension.pdf` | Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts | [arxiv:2306.04723](https://arxiv.org/abs/2306.04723) |
| 2024 | `2024_traub_augrc.pdf` | Overcoming Common Flaws in the Evaluation of Selective Classification Systems | [arxiv:2407.01032](https://arxiv.org/abs/2407.01032) |

## ICML (5)

| year | file | title | source |
|---|---|---|---|
| 2020 | `2020_moon_crl.pdf` | Confidence-Aware Learning for Deep Neural Networks | [arxiv:2007.01458](https://arxiv.org/abs/2007.01458) |
| 2020 | `2020_vanamersfoort_duq.pdf` | Uncertainty Estimation Using a Single Deep Deterministic Neural Network | [arxiv:2003.02037](https://arxiv.org/abs/2003.02037) |
| 2022 | `2022_hendrycks_maxlogit.pdf` | Scaling Out-of-Distribution Detection for Real-World Settings | [arxiv:1911.11132](https://arxiv.org/abs/1911.11132) |
| 2022 | `2022_sun_knn_ood.pdf` | Out-of-Distribution Detection with Deep Nearest Neighbors | [arxiv:2204.06507](https://arxiv.org/abs/2204.06507) |
| 2025 | `2025_skean_layer_by_layer.pdf` | Layer by Layer: Uncovering Hidden Representations in Language Models | [arxiv:2502.02013](https://arxiv.org/abs/2502.02013) |

## AAAI (1)

| year | file | title | source |
|---|---|---|---|
| 2021 | `2021_podolskiy_mahalanobis_transformers.pdf` | Revisiting Mahalanobis Distance for Transformer-Based Out-of-Domain Detection | [arxiv:2101.03778](https://arxiv.org/abs/2101.03778) |

## CVPR (3)

| year | file | title | source |
|---|---|---|---|
| 2022 | `2022_wang_vim.pdf` | ViM: Out-Of-Distribution with Virtual-logit Matching | [arxiv:2203.10807](https://arxiv.org/abs/2203.10807) |
| 2023 | `2023_mukhoti_ddu.pdf` | Deep Deterministic Uncertainty: A New Simple Baseline | [arxiv:2102.11582](https://arxiv.org/abs/2102.11582) |
| 2023 | `2023_zhu_openmix.pdf` | OpenMix: Exploring Outlier Samples for Misclassification Detection | [arxiv:2303.17093](https://arxiv.org/abs/2303.17093) |

## TPAMI (1)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_zhu_revisiting_confidence_failure_prediction.pdf` | Revisiting Confidence Estimation: Towards Reliable Failure Prediction | [arxiv:2403.02886](https://arxiv.org/abs/2403.02886) |

## TMLR (4)

| year | file | title | source |
|---|---|---|---|
| 2022 | `2022_rabanser_training_dynamics.pdf` | Selective Prediction via Training Dynamics | [arxiv:2205.13532](https://arxiv.org/abs/2205.13532) |
| 2023 | `2023_lahlou_deup.pdf` | DEUP: Direct Epistemic Uncertainty Prediction | [arxiv:2102.08501](https://arxiv.org/abs/2102.08501) |
| 2024 | `2024_lang_ood_nlp_survey.pdf` | A Survey on Out-of-Distribution Detection in NLP | [arxiv:2305.03236](https://arxiv.org/abs/2305.03236) |
| 2024 | `2024_lin_generating_with_confidence.pdf` | Generating with Confidence: Uncertainty Quantification for Black-box LLMs | [arxiv:2305.19187](https://arxiv.org/abs/2305.19187) |

## COLM (1)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_marks_geometry_of_truth.pdf` | The Geometry of Truth: Emergent Linear Structure in LLM Representations of True/False Datasets | [arxiv:2310.06824](https://arxiv.org/abs/2310.06824) |

## UAI (1)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_cattelan_fix_broken_confidence.pdf` | How to Fix a Broken Confidence Estimator | [arxiv:2305.15508](https://arxiv.org/abs/2305.15508) |

## KDD (1)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_snyder_early_detection_hallucination.pdf` | On Early Detection of Hallucinations in Factual Question Answering | [arxiv:2312.14183](https://arxiv.org/abs/2312.14183) |

## IJCV (1)

| year | file | title | source |
|---|---|---|---|
| 2024 | `2024_yang_generalized_ood_survey.pdf` | Generalized Out-of-Distribution Detection: A Survey | [arxiv:2110.11334](https://arxiv.org/abs/2110.11334) |

## arXiv (12)

| year | file | title | source |
|---|---|---|---|
| 2021 | `2021_ren_relative_mahalanobis.pdf` | A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection | [arxiv:2106.09022](https://arxiv.org/abs/2106.09022) |
| 2022 | `2022_kadavath_mostly_know.pdf` | Language Models (Mostly) Know What They Know | [arxiv:2207.05221](https://arxiv.org/abs/2207.05221) |
| 2023 | `2023_belrose_tuned_lens.pdf` | Eliciting Latent Predictions from Transformers with the Tuned Lens | [arxiv:2303.08112](https://arxiv.org/abs/2303.08112) |
| 2024 | `2024_kossen_semantic_entropy_probes.pdf` | Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs | [arxiv:2406.15927](https://arxiv.org/abs/2406.15927) |
| 2024 | `2024_liu_supervised_ue_llm.pdf` | Uncertainty Estimation and Quantification for LLMs: A Simple Supervised Approach | [arxiv:2404.15993](https://arxiv.org/abs/2404.15993) |
| 2025 | `2025_clue_hidden_state_clustering.pdf` | CLUE: Non-parametric Verification from Experience via Hidden-State Clustering | [arxiv:2510.01591](https://arxiv.org/abs/2510.01591) |
| 2025 | `2025_kim_layerwise_inference_dynamics.pdf` | On the Effect of Uncertainty on Layer-wise Inference Dynamics | [arxiv:2507.06722](https://arxiv.org/abs/2507.06722) |
| 2025 | `2025_sridhar_clinical_doubt.pdf` | Mapping Clinical Doubt: Locating Linguistic Uncertainty in LLMs | [arxiv:2511.22402](https://arxiv.org/abs/2511.22402) |
| 2025 | `2025_suresh_clap_cross_layer_probing.pdf` | Cross-Layer Attention Probing for Fine-Grained Hallucination Detection | [arxiv:2509.09700](https://arxiv.org/abs/2509.09700) |
| 2026 | `2026_cho_confidence_manifold.pdf` | The Confidence Manifold: Geometric Structure of Correctness | [arxiv:2602.08159](https://arxiv.org/abs/2602.08159) |
| 2026 | `2026_delajara_representation_trajectories.pdf` | Representation Trajectories Matter | [arxiv:2607.26565](https://arxiv.org/abs/2607.26565) |
| 2026 | `2026_eusebi_calibrated_uncertainty_trajectories.pdf` | Reading Calibrated Uncertainty from Language Model Trajectories | [arxiv:2605.22864](https://arxiv.org/abs/2605.22864) |

## Templates

| dir | venue | source |
|---|---|---|
| `templates/acl-style-files` | ACL / NAACL / EMNLP / EACL (shared ACL style, `acl.sty`, `acl_natbib.bst`) | github.com/acl-org/acl-style-files |
| `templates/neurips2025` | NeurIPS 2025 | media.neurips.cc |
| `templates/iclr2026` | ICLR 2026 | github.com/ICLR/Master-Template |
| `templates/icml2025` | ICML 2025 | media.icml.cc |
| `templates/aaai2026-quarto/_extensions/aaai2026` | AAAI-26 (`aaai2026.sty`, `aaai2026.bst`; aaai.org blocks direct download, this is a mirror of the kit's style files) | github.com/Selbosh/aaai2026-quarto |
| `templates/ieeetran` | IEEE Transactions / IEEE Access (`IEEEtran.cls`) | CTAN |