mlsanityData to 13 Oct 2026

CaLM: Contrasting Large and Small Language Models to Verify Grounded Generation

arXiv 2406.05365 ↗First posted 8 Jun 2024cs.CL cs.AI cs.LGcited by 11

Not on the map. Why?

Abstract (arXiv)

Grounded generation aims to equip language models (LMs) with the ability to produce more credible and accountable responses by accurately citing verifiable sources. However, existing methods, by either feeding LMs with raw or preprocessed materials, remain prone to errors. To address this, we introduce CaLM, a novel verification framework. CaLM leverages the insight that a robust grounded response should be consistent with information derived solely from its cited sources. Our framework empowers smaller LMs, which rely less on parametric memory and excel at processing relevant information given a query, to validate the output of larger LMs. Larger LM responses that closely align with the smaller LMs' output, which relies exclusively on cited documents, are verified. Responses showing discrepancies are iteratively refined through a feedback loop. Experiments on three open-domain question-answering datasets demonstrate significant performance gains of 1.5% to 7% absolute average without any required model fine-tuning.

Key facts. Cited by 11 papers in the map (data through 13 Oct 2026). Builds on 26 of its 40 references that are in the map. First posted on arXiv 8 Jun 2024.

Builds on

26 of 40 references.

References · 26

  1. 2022Training language models to follow instructions with human feedbackcited by 15,856
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,492
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,831
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,657
  5. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,797
  6. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,412
  7. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,736
  8. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,205
  9. 2023PaLM 2 Technical Reportcited by 1,162
  10. 2023Bias and Fairness in Large Language Models: A Surveycited by 636
  11. 2023Enabling Large Language Models to Generate Text with Citationscited by 588
  12. 2019ELI5: Long Form Question Answeringcited by 571
14 more
  1. 2020KILT: a Benchmark for Knowledge Intensive Language Taskscited by 571
  2. 2023LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusioncited by 553
  3. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 468
  4. 2021Large Dual Encoders Are Generalizable Retrieverscited by 455
  5. 2020AmbigQA: Answering Ambiguous Open-domain Questionscited by 386
  6. 2023Evaluating Verifiability in Generative Search Enginescited by 299
  7. 2022Teaching language models to support answers with verified quotescited by 291
  8. 2024R-Judge: Benchmarking Safety Risk Awareness for LLM Agentscited by 280
  9. 2021Measuring Attribution in Natural Language Generation Modelscited by 253
  10. 2023Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2cited by 229
  11. 2023Universal Self-Consistency for Large Language Model Generationcited by 200
  12. 2022Rethinking with Retrieval: Faithful Large Language Model Inferencecited by 145
  13. 2021The Web Is Your Oyster - Knowledge-Intensive NLP against a Very Large Web Corpuscited by 60
  14. 2023LLatrieval: LLM-Verified Retrieval for Verifiable Generationcited by 52

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2024: 2; 2025: 5; 2026: 4.

11 papers.

  1. 2024What is the Role of Small Models in the LLM Era: A Surveycited by 41
  2. 2024Measuring and Enhancing Trustworthiness of LLMs in RAG through Grounded Attributions and Learning to Refusecited by 33
  3. 2025CiteFix: Enhancing RAG Accuracy Through Post-Processing Citation Correctioncited by 8
  4. 2025Efficient LLM Collaboration via Planningcited by 7
  5. 2025A Survey on Collaborating Small and Large Language Models for Performance, Cost-effectiveness, Cloud-edge Privacy, and Trustworthinesscited by 6
  6. 2025TROVE: A Challenge for Fine-Grained Text Provenance via Source Sentence Tracing and Relationship Classificationcited by 4
  7. 2025SNaRe: Domain-aware Data Generation for Low-Resource Event Detectioncited by 2
  8. 2026GenProve: Learning to Generate Text with Fine-Grained Provenancecited by 2
  9. 2026Structuring the Space of Sociotechnical Alignmentcited by 2
  10. 2026Attribution Techniques for Mitigating Hallucinated Information in RAG Systems: A Surveycited by 0
  11. 2026Train, Retrieve, or Both? A Four-Arm Head-to-Head for Correct Statutory Citation on the Ontario Residential Tenancies Actcited by 0