# VeriFastScore: Speeding up long-form factuality evaluation

arXiv: https://arxiv.org/abs/2505.16973 · HTML: https://mlsanity.com/paper/2505.16973/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 29 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 15 of 15)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://mlsanity.com/paper/2305.14251/) (2023)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (2021)
- [FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation](https://mlsanity.com/paper/2310.03214/) (2023)
- [MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documents](https://mlsanity.com/paper/2404.10774/) (2024)
- [FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarios](https://mlsanity.com/paper/2307.13528/) (2023)
- [VERISCORE: Evaluating the factuality of verifiable claims in long-form text generation](https://mlsanity.com/paper/2406.19276/) (2024)
- [A Critical Evaluation of Evaluations for Long-form Question Answering](https://mlsanity.com/paper/2305.18201/) (2023)
- [Factcheck-Bench: Fine-Grained Evaluation Benchmark for Automatic Fact-checkers](https://mlsanity.com/paper/2311.09000/) (2023)
- [A Closer Look at Claim Decomposition](https://mlsanity.com/paper/2403.11903/) (2024)
- [Decomposition Dilemmas: Does Claim Decomposition Boost or Burden Fact-Checking Performance?](https://mlsanity.com/paper/2411.02400/) (2024)
- [Retrieval Augmented Fact Verification by Synthesizing Contrastive Arguments](https://mlsanity.com/paper/2406.09815/) (2024)
- [Truth or Mirage? Towards End-to-End Factuality Evaluation with LLM-Oasis](https://mlsanity.com/paper/2411.19655/) (2024)
- [Towards Long Context Hallucination Detection](https://mlsanity.com/paper/2504.19457/) (2025)

## Built on by (most cited first, 8 of 8)

- [LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scale](https://mlsanity.com/paper/2603.24080/) (2026)
- [Distill and Align Decomposition for Enhanced Claim Verification](https://mlsanity.com/paper/2602.21857/) (2026)
- [Probabilistic Guarantees for Reducing Contextual Hallucinations in LLMs](https://mlsanity.com/paper/2601.00641/) (2026)
- [MIRAGE: Defending Long-Form RAG Against Misinformation Pollution](https://mlsanity.com/paper/2607.05069/) (2026)
- [Symbolic Augmentation Closes a Canonical-Equivalence Blind Spot in Neural Fact-Checkers](https://mlsanity.com/paper/2607.16212/) (2026)
- [LLMPEDIA: Browsing, Verifying, and Comparing the Parametric Encyclopedic Knowledge of LLMs](https://mlsanity.com/paper/2609.01182/) (2026)
- [FrankenReport: Early Exiting in Long-Form Generation Using Expected Value of Computation](https://mlsanity.com/paper/2609.06320/) (2026)
- [EAVer: Long-Form Factuality Verification as an End-to-End Agentic Policy](https://mlsanity.com/paper/2609.22223/) (2026)

## Abstract (arXiv)

Metrics like FactScore and VeriScore that evaluate long-form factuality operate by decomposing an input response into atomic claims and then individually verifying each claim. While effective and interpretable, these methods incur numerous LLM calls and can take upwards of 100 seconds to evaluate a single response, limiting their practicality in large-scale evaluation and training scenarios. To address this, we propose VeriFastScore, which leverages synthetic data to fine-tune Llama3.1 8B for simultaneously extracting and verifying all verifiable claims within a given text based on evidence from Google Search. We show that this task cannot be solved via few-shot prompting with closed LLMs due to its complexity: the model receives ~4K tokens of evidence on average and needs to concurrently decompose claims, judge their verifiability, and verify them against noisy evidence. However, our fine-tuned VeriFastScore model demonstrates strong correlation with the original VeriScore pipeline at both the example level (r=0.80) and system level (r=0.94) while achieving an overall speedup of 6.6x (9.9x excluding evidence retrieval) over VeriScore. To facilitate future factuality research, we publicly release our VeriFastScore model and synthetic datasets.