mlsanityData to 7 Oct 2026

Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answering

arXiv 2307.16877 ↗First posted 31 Jul 2023cs.CL cs.AIcited by 150

Not on the map. Why?

Cited by 31 in Retrieval-augmented generation, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Retriever-augmented instruction-following models are attractive alternatives to fine-tuned approaches for information-seeking tasks such as question answering (QA). By simply prepending retrieved documents in its input along with an instruction, these models can be adapted to various information domains and tasks without additional fine-tuning. While the model responses tend to be natural and fluent, the additional verbosity makes traditional QA evaluation metrics such as exact match (EM) and F1 unreliable for accurately quantifying model performance. In this work, we investigate the performance of instruction-following models across three information-seeking QA tasks. We use both automatic and human evaluation to evaluate these models along two dimensions: 1) how well they satisfy the user's information need (correctness), and 2) whether they produce a response based on the provided knowledge (faithfulness). Guided by human evaluation and analysis, we highlight the shortcomings of traditional metrics for both correctness and faithfulness. We then propose simple token-overlap based and model-based metrics that reflect the true performance of these models. Our analysis reveals that instruction-following models are competitive, and sometimes even outperform fine-tuned models for correctness. However, these models struggle to stick to the provided knowledge and often hallucinate in their responses. We hope our work encourages a more holistic evaluation of instruction-following models for QA. Our code and data is available at https://github.com/McGill-NLP/instruct-qa

Key facts. Cited by 150 papers in the map (data through 7 Oct 2026). Builds on 66 of its 91 references that are in the map. First posted on arXiv 31 Jul 2023.

Builds on

66 of 91 references.

References · 66

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2022Training language models to follow instructions with human feedbackcited by 15,712
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  7. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  8. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  9. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,115
  10. 2017Deep reinforcement learning from human preferencescited by 3,846
  11. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  12. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
38 more
  1. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  2. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  5. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  6. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  7. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  8. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  9. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,462
  10. 2021Calibrate Before Use: Improving Few-Shot Performance of Language Modelscited by 1,371
  11. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
  12. 2021Scaling Language Models: Methods, Analysis & Insights from Training Gophercited by 1,199
  13. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,103
  14. 2023Large Language Models are not Fair Evaluatorscited by 1,050
  15. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 940
  16. 2022Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Taskscited by 858
  17. 2023REPLUG: Retrieval-Augmented Black-Box Language Modelscited by 786
  18. 2017Reading Wikipedia to Answer Open-Domain Questionscited by 777
  19. 2023Large Language Modelscited by 774
  20. 2022Teaching Models to Express Their Uncertainty in Wordscited by 731
  21. 2023Can Large Language Models Be an Alternative to Human Evaluations?cited by 729
  22. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 696
  23. 2019Latent Retrieval for Weakly Supervised Open Domain Question Answeringcited by 691
  24. 2021Retrieval Augmentation Reduces Hallucination in Conversationcited by 691
  25. 2023Instruction Tuning with GPT-4cited by 676
  26. 2018CoQA: A Conversational Question Answering Challengecited by 674
  27. 2023Faith and Fate: Limits of Transformers on Compositionalitycited by 496
  28. 2022News Summarization and Evaluation in the Era of GPT-3cited by 324
  29. 2021Measuring Attribution in Natural Language Generation Modelscited by 253
  30. 2022OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalizationcited by 235
  31. 2020Reducing conversational agents' overconfidence through linguistic calibrationcited by 209
  32. 2020Answering Complex Open-Domain Questions with Multi-Hop Dense Retrievalcited by 160
  33. 2019Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answeringcited by 147
  34. 2020Open-Domain Question Answering Goes Conversational via Question Rewritingcited by 147
  35. 2022On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models?cited by 137
  36. 2021Q2: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answeringcited by 131
  37. 2022Internet-augmented language models through few-shot prompting for open-domain question answeringcited by 131
  38. 2017Simple and Effective Multi-Paragraph Reading Comprehensioncited by 120

50 most cited of 66 All 66 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 15; 2024: 71; 2025: 42; 2026: 22.

150 papers.

Sort:

By area:

  1. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  2. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,582
  3. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 695
  4. 2023Instruction Tuning for Large Language Models: A Surveycited by 617
  5. 2023Benchmarking Large Language Models in Retrieval-Augmented Generationcited by 307
  6. 2024MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documentscited by 251
  7. 2024RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systemscited by 115
  8. 2023LawBench: Benchmarking Legal Knowledge of Large Language Modelscited by 111
  9. 2024Large Language Models Meet NLP: A Surveycited by 102
  10. 2024Tug-of-War Between Knowledge: Exploring and Resolving Knowledge Conflicts in Retrieval-Augmented Language Modelscited by 93
  11. 2024Cutting Off the Head Ends the Conflict: A Mechanism for Interpreting and Mitigating Knowledge Conflicts in Language Modelscited by 87
  12. 2024Fast Adversarial Attacks on Language Models In One GPU Minutecited by 85
38 more
  1. 2024FaithEval: Can Your Language Model Stay Faithful to Context, Even If "The Moon is Made of Marshmallows"cited by 85
  2. 2023Cognitive Mirage: A Review of Hallucinations in Large Language Modelscited by 83
  3. 2024RAG vs Fine-tuning: Pipelines, Tradeoffs, and a Case Study on Agriculturecited by 80
  4. 2024Retrieval-Augmented Generation for Natural Language Processing: A Surveycited by 70
  5. 2024LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Playcited by 65
  6. 2023On Early Detection of Hallucinations in Factual Question Answeringcited by 54
  7. 2024Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?cited by 53
  8. 2024Generate-then-Ground in Retrieval-Augmented Generation for Multi-hop Question Answeringcited by 51
  9. 2025MTRAG: A Multi-Turn Conversational Benchmark for Evaluating Retrieval-Augmented Generation Systemscited by 50
  10. 2024Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Eracited by 45
  11. 2024Improving fine-grained understanding in image-text pre-trainingcited by 44
  12. 2024Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Ratescited by 42
  13. 2023Retrieval-augmented Generation to Improve Math Question-Answering: Trade-offs Between Groundedness and Human Preferencecited by 37
  14. 2024AI Hallucinations: A Misnomer Worth Clarifyingcited by 34
  15. 2024Genie: Achieving Human Parity in Content-Grounded Datasets Generationcited by 31
  16. 2024RepLiQA: A Question-Answering Dataset for Benchmarking LLMs on Unseen Reference Contentcited by 26
  17. 2024Beyond the Turn-Based Game: Enabling Real-Time Conversations with Duplex Modelscited by 26
  18. 2024Automated Evaluation of Retrieval-Augmented Language Models with Task-Specific Exam Generationcited by 25
  19. 2024RAG-RewardBench: Benchmarking Reward Models in Retrieval Augmented Generation for Preference Alignmentcited by 22
  20. 2025Brains vs. Bytes: Evaluating LLM Proficiency in Olympiad Mathematicscited by 22
  21. 2024Generative AI in EU Law: Liability, Privacy, Intellectual Property, and Cybersecuritycited by 21
  22. 2024AmbigDocs: Reasoning across Documents on Different Entities under the Same Namecited by 21
  23. 2024ArabicaQA: A Comprehensive Dataset for Arabic Question Answeringcited by 20
  24. 2024KGARevion: An AI Agent for Knowledge-Intensive Biomedical QAcited by 20
  25. 2025Robust Hallucination Detection in LLMs via Adaptive Token Selectioncited by 19
  26. 2023User Modeling in the Era of Large Language Models: Current Research and Future Directionscited by 18
  27. 2024GraphOTTER: Evolving LLM-based Graph Reasoning for Complex Table Question Answeringcited by 18
  28. 2024LRP4RAG: Detecting Hallucinations in Retrieval-Augmented Generation via Layer-wise Relevance Propagationcited by 17
  29. 2024Truth-Aware Context Selection: Mitigating Hallucinations of Large Language Models Being Misled by Untruthful Contextscited by 15
  30. 2024Investigating and Addressing Hallucinations of LLMs in Tasks Involving Negationcited by 15
  31. 2024Writing Style Matters: An Examination of Bias and Fairness in Information Retrieval Systemscited by 15
  32. 2024On Subjective Uncertainty Quantification and Calibration in Natural Language Generationcited by 14
  33. 2025Refining Input Guardrails: Enhancing LLM-as-a-Judge Efficiency Through Chain-of-Thought Fine-Tuning and Alignmentcited by 14
  34. 2023Bias A-head? Analyzing Bias in Transformer-Based Language Model Attention Headscited by 13
  35. 2024Faithful Temporal Question Answering over Heterogeneous Sourcescited by 13
  36. 2024SECURE: Benchmarking Large Language Models for Cybersecuritycited by 13
  37. 2024From Loops to Oops: Fallback Behaviors of Language Models Under Uncertaintycited by 13
  38. 2024Multilingual Retrieval Augmented Generation for Culturally-Sensitive Tasks: A Benchmark for Cross-lingual Robustnesscited by 13

50 most cited of 150 All 150 (JSON)