Factual Consistency Evaluation

Evaluating scientific arguments requires assessing the strict consistency between a claim and its underlying multimodal evidence. However, existing benchmarks lack the scale, domain diversity, and visual complexity needed to evaluate this alignment realistically. To address this gap, we introduce M2-Verify, a large-scale multimodal dataset for checking scientific claim consistency. Sourced from PubMed and arXiv, M2-Verify provides over 469K instances across 16 domains, rigorously validated through expert audits. Extensive baseline experiments show that state-of-the-art models struggle to maintain robust consistency. While top models achieve up to 85.8% Micro-F1 on low-complexity medical perturbations, performance drops to 61.6% on high-complexity challenges like anatomical shifts. Furthermore, expert evaluations expose hallucinations when models generate scientific explanations for their alignment decisions. Finally, we demonstrate our dataset's utility and provide comprehensive usage guidelines.

Large language model (LLM)-based persona agents are rapidly being adopted as scalable proxies for human participants across diverse domains. Yet there is no systematic method for verifying whether a persona agent's responses remain free of contradictions and factual inaccuracies throughout an interaction. A principle from interrogation methodology offers a lens: no matter how elaborate a fabricated identity, systematic interrogation will expose its contradictions. We apply this principle to propose PICon, an evaluation framework that probes persona agents through logically chained multi-turn questioning. PICon evaluates consistency along three core dimensions: internal consistency (freedom from self-contradiction), external consistency (alignment with real-world facts), and retest consistency (stability under repetition). Evaluating seven groups of persona agents alongside 63 real human participants, we find that even systems previously reported as highly consistent fail to meet the human baseline across all three dimensions, revealing contradictions and evasive responses under chained questioning. This work provides both a conceptual foundation and a practical methodology for evaluating persona agents before trusting them as substitutes for human participants. We provide the source code and an interactive demo at: https://kaist-edlab.github.io/picon/

Large language models (LLMs) are now widely used as judges, yet their decisions can change under presentation choices that should be irrelevant. We study one such source of instability: candidate-order sensitivity in listwise factuality evaluation, where several answers can look similarly polished while differing substantially in hallucination risk. We introduce PCFJudge, an inference-time method that reruns the same factuality-first listwise prompt over multiple orderings of the same candidate set and aggregates the resulting scores, ranks, and uncertainty signals into a single consensus decision. On RewardBench 2 Factuality, the final seven-permutation aggregate (K=7) improves top-1 selection accuracy from 86.00% to 91.33% with GPT-5.4 and from 86.33% to 89.67% with Claude Sonnet 4.6. These results suggest that candidate order can be a meaningful source of factuality-judging error and that marginalizing over this nuisance variation can improve the reliability of LLM evaluation.

Evidence-grounded generation produces summaries whose claims should be supported by supplied evidence, but claim-level verifiers provide noisy feedback and can reward models that simply say less. We study this problem in clinical Brief Hospital Course summarization, where outputs must remain grounded in patient-specific EHR evidence. We introduce VERI-DPO, a preference-mining framework that converts noisy claim verification into coverage-controlled summary-level preferences. For each evidence-window prompt, VERI-DPO samples multiple candidate summaries, decomposes them into claims, verifies each claim against patient evidence, and forms a preference pair only when the chosen summary has better aggregate verifier-estimated support while retaining comparable verifiable content. Standard Direct Preference Optimization then distills these pairs into a single-sample policy, avoiding inference-time reranking. On patient-disjoint MIMIC-III-Ext-VeriFact-BHC test data, VERI-DPO reduces Not Supported rates from 10.7% to 1.9% under the mining verifier and from 11.6% to 6.4% under a separately prompted GPT-4o judge. In 100 blinded pairwise assessments by two domain researchers, VERI-DPO is preferred over the base model 56 times versus 18 times for factual faithfulness. In a locked zero-shot MIMIC-IV-Ext-BHC transfer test with 1,000 patients and no model adaptation, VERI-DPO lowers Not Supported rates with nearly unchanged scored-claim counts. Multi-seed ablations show that verifier-guided pair construction drives the gains, while coverage and anti-degeneration controls prevent apparent factuality improvements from coming from shorter or less checkable outputs.

We present a configurable pipeline for generating multilingual sets of entities with specified characteristics, such as domain, geographical location and popularity, using data from Wikipedia and Wikidata. These datasets are intended for evaluating the factuality of LLMs' long-form generation, thereby complementing evaluation based on short-form QA datasets. We present the RiDiC dataset as an example of this approach. RiDiC contains 3,000 entities from three domains -- rivers, natural disasters, and car models -- spanning different popularity tiers. Each entity is accompanied by its geographical location, English and Chinese names (if available) and relevant English and Chinese Wikipedia content, which is used to evaluate LLMs' responses. Generations about RiDiC entities were obtained from three LLMs in English and Chinese. These were then evaluated using a third-party factuality checker, which showed that entities from our dataset caused even frontier models to hallucinate. To facilitate the evaluation of LLMs' long-form factuality in multiple languages, the code, data, and generation/evaluation scripts have been released.

Search-augmented LLM agents can produce deep research reports (DRRs), but verifying claim-level factuality remains challenging. Existing fact-checkers are primarily designed for general-domain, factoid-style atomic claims, and there is no benchmark to test whether such verifiers transfer to DRRs. Yet building such a benchmark is itself difficult. We first show that static expert-labeled benchmarks are brittle in this setting: in a controlled study with PhD-level specialists, unassisted experts achieve only 60.8% accuracy on a hidden micro-gold set of verifiable claims. We propose Evolving Benchmarking via Audit-then-Score (AtS), where benchmark labels and rationales are explicitly revisable: when a verifier disagrees with the current benchmark, it must submit evidence; an auditor adjudicates the dispute; and accepted revisions update the benchmark before models are scored. Across four AtS rounds, expert micro-gold accuracy rises to 90.9%, indicating experts are substantially more reliable as auditors than as one-shot labelers. We instantiate AtS as DeepFact-Bench, a versioned DRR factuality benchmark with auditable rationales, and DeepFact-Eval, a document-level verification agent (with a grouped lite variant) that outperforms existing verifiers on DeepFact-Bench and transfers well to external factuality datasets.

Uncertainty quantification has emerged as an effective approach to closed-book hallucination detection for LLMs, but existing methods are largely designed for short-form outputs and do not generalize well to long-form generation. We introduce a taxonomy for fine-grained uncertainty quantification in long-form LLM outputs that distinguishes methods by design choices at three stages: response decomposition, unit-level scoring, and response-level aggregation. We formalize several families of consistency-based black-box scorers, providing generalizations and extensions of existing methods. We also introduce FactScore-STEM-Geo, a new 400-question long-form QA dataset spanning four categories across STEM and Geography. In our experiments across multiple LLMs and datasets, we find 1) claim-response entailment consistently performs better or on par with more complex claim-level scorers, 2) claim-level scoring generally yields better results than sentence-level scoring, and 3) uncertainty-aware decoding is highly effective for improving the factuality of long-form outputs. Our framework clarifies relationships between prior methods, enables apples-to-apples comparisons, and provides practical guidance for selecting components for fine-grained UQ.

Evaluating factual consistency is essential for reliable text summarization, particularly in high-stakes domains such as healthcare and news. However, most existing evaluation metrics overlook Bangla, a widely spoken yet under-resourced language, and often depend on reference summaries. We introduce BanglaSummEval, a reference-free, question-answering-based framework for evaluating factual consistency in Bangla summarization. The proposed method assesses both factual accuracy and content coverage through automatically generated questions and answers derived from the source document and the summary. A single multilingual instruction-tuned language model handles question generation, question answering, candidate answer extraction, and question importance weighting. This unified design reduces system complexity and computational cost. To capture semantic consistency beyond surface-level overlap, we use BERTScore-Recall for answer comparison. We validate BanglaSummEval on 300 human-written summaries from educational and medical domains, demonstrating strong correlation with expert human judgments (Pearson's $r = 0.694$, Spearman's $ρ= 0.763$). By providing interpretable, step-wise diagnostics alongside reliable evaluation scores, BanglaSummEval offers a practical and transparent solution for factual consistency evaluation in low-resource language settings.

Hallucination in large language models (LLMs) remains an acute concern, contributing to the spread of misinformation and diminished public trust, particularly in high-risk domains. Among hallucination types, factuality is crucial, as it concerns a model's alignment with established world knowledge. Adversarial factuality, defined as the deliberate insertion of misinformation into prompts with varying levels of expressed confidence, tests a model's ability to detect and resist confidently framed falsehoods. Existing work lacks high-quality, domain-specific resources for assessing model robustness under such adversarial conditions, and no prior research has examined the impact of injected misinformation on long-form text factuality. To address this gap, we introduce AdversaRiskQA, the first verified and reliable benchmark systematically evaluating adversarial factuality across Health, Finance, and Law. The benchmark includes two difficulty levels to test LLMs' defensive capabilities across varying knowledge depths. We propose two automated methods for evaluating the adversarial attack success and long-form factuality. We evaluate six open- and closed-source LLMs from the Qwen, GPT-OSS, and GPT families, measuring misinformation detection rates. Long-form factuality is assessed on Qwen3 (30B) under both baseline and adversarial conditions. Results show that after excluding meaningless responses, Qwen3 (80B) achieves the highest average accuracy, while GPT-5 maintains consistently high accuracy. Performance scales non-linearly with model size, varies by domains, and gaps between difficulty levels narrow as models grow. Long-form evaluation reveals no significant correlation between injected misinformation and the model's factual output. AdversaRiskQA provides a valuable benchmark for pinpointing LLM weaknesses and developing more reliable models for high-stakes applications.

NLP systems increasingly aggregate claims from retrieved passages, documents, and generated reasoning steps. Yet current factuality methods mostly verify claims individually or compare pairs of claims, leaving open the question of whether the entire aggregated claim set can be retained coherently. This misses a fundamental failure mode: claims may be locally plausible and pairwise compatible while still being jointly inconsistent. We formalize global consistency repair: given an inconsistent natural-language claim set, preserve the largest coherent subset while localizing the compact sources of inconsistency. We introduce QXR, an adaptive diagnosis-and-repair framework that treats an LLM as a noisy set-level consistency judge, localizes minimal conflicting subsets, and repairs the claim set by deleting a small hitting set of claims. Across LLM judges and benchmarks derived from VitaminC, FEVER, and conflicting-evidence RAG, QXR preserves substantially more valid claims than direct all-at-once judging while maintaining high precision. These results suggest that factuality in claim aggregation systems should be treated as a set-level repair problem, not only as local verification.

Large language models (LLMs) now support contexts of up to 1M tokens, but their strengths and weaknesses on complex long-context tasks remain unclear. To study this, we focus on multi-document legal case summarization, where a single case often spans many documents exceeding 100K tokens. We systematically evaluate 12 frontier LLMs with Gavel, which consists of Gavel-Ref, a reference-based evaluation framework with checklist, residual-fact, and writing-style evaluations, and Gavel-Agent, a reference-free agent for evaluating factual coverage directly from source documents. Our results show that current models are more prone to omitting key information than hallucinating. They all perform well on simple checklist items, such as filing date, but struggle with rare and complex items, such as settlements. Performance also declines as case length increases. To meta-evaluate Gavel, we collect 160 hours of human annotations. Gavel-Agent reduces token usage by at least 36% compared to end-to-end and chunk-by-chunk methods while achieving competitive performance. Gavel-Agent also generalizes to the medical domain, performing the best with at least 77% fewer tokens.

LLMs are remarkable artifacts that have revolutionized a range of knowledge-intensive tasks. A significant contributor is their factual knowledge, which, to date, remains poorly understood, and is usually analyzed from biased samples. In this paper, we provide a framework and the results of a multi-dimensional analysis of GPTKB v1.5 (Hu et al., 2025a), a recursively elicited Knowledge Base (KB) of 100 million facts (or beliefs) of a frontier LLM, namely, GPT-4.1. Given the scale of the elicited facts, we provide a multi-dimensional approach to qualitatively and quantitatively analyze these facts as opposed to the mainstream fact completion benchmarks, which are prone to availability bias. We find that the models' factual knowledge differs quite significantly from established knowledge bases, and that its accuracy is significantly lower than indicated by previous benchmarks. We also find that inconsistency, ambiguity and hallucinations are major issues, shedding light on future research opportunities in neuro-symbolic AI concerning extraction, consolidation and verification of factual LLM knowledge.

The rapid adoption of Large Language Models (LLMs) has raised important concerns about the factual reliability of their outputs, particularly in low-resource languages such as Urdu. Existing automated fact-checking systems are predominantly developed for English, leaving a significant gap for the more than 200 million Urdu speakers worldwide. In this work, we present UrduFactBench and UrduFactQA, two novel hand-annotated benchmarks designed to enable fact-checking and factual consistency evaluation in Urdu. While UrduFactBench focuses on claim verification, UrduFactQA targets the factuality of LLMs in question answering. These resources, the first of their kind for Urdu, were developed through a multi-stage annotation process involving native Urdu speakers. To complement these benchmarks, we introduce UrduFactCheck, a modular fact-checking framework that incorporates both monolingual and translation-based evidence retrieval strategies to mitigate the scarcity of high-quality Urdu evidence. Leveraging these resources, we conduct an extensive evaluation of twelve LLMs and demonstrate that translation-augmented pipelines consistently enhance performance compared to monolingual ones. Our findings reveal persistent challenges for open-source LLMs in Urdu and underscore the importance of developing targeted resources. All code and data are publicly available at https://github.com/mbzuai-nlp/UrduFactCheck.

As Large Language Models (LLMs) become increasingly used for question-answering (QA), relying on static, pre-annotated references for evaluation poses significant challenges in cost, scalability, and completeness. Meanwhile, using LLMs themselves as evaluators without external grounding remains unreliable for objective tasks, as they systematically over-accept incorrect answers, fabricate supporting rationales, and degrade sharply on questions that fall outside their training data. We propose Search-AuGmented Evaluation (SAGE), a framework to assess LLM outputs without fixed ground-truth answers. Unlike conventional metrics that compare to static references or depend solely on LLM-as-a-judge knowledge, SAGE acts as an agent that actively retrieves and synthesizes external evidence. It iteratively generates web queries, collects information, summarizes findings, and refines subsequent searches through reflection. By reducing dependence on static reference-driven evaluation protocols, SAGE offers a scalable and adaptive alternative for evaluating the factuality of LLMs. Experimental results on multiple free-form QA benchmarks show that SAGE achieves substantial to perfect agreement with human evaluations.