KnowHal: A Knowledge-Driven Benchmark for Comprehensive Multimodal Hallucination Evaluation
Score
78首次将知识幻觉纳入统一多模态评估框架，填补现有基准的维度空白。
样本构建结合自动化与人工验证，实验规模覆盖14个模型，分析细致。
为MLLM开发者提供可直接使用的评测基准，有助于改进模型可靠性与鲁棒性。
针对当前多模态大模型可信度热点问题，知识幻觉评估需求迫切。
来自arXiv论文，方法流程透明，数据构建过程可追溯。
Base 86.5 × T2 Source weight ×0.9 = 78
Curation line 70Selected
Why selected
The paper is rigorous and comprehensive, addressing a real pain point with a reproducible benchmark, making a clear and practical contribution.
AI commentary
「This paper is worth attention because it formally brings knowledge hallucination into the mainstream of multimodal evaluation, instead of leaving knowledge errors scattered across small datasets. For teams building MLLM benchmarks or improving model trustworthiness, these 1,800 paired samples are a great starting point; for general researchers, the negative-question design reminds us that models tend to be too compliant with false premises.」
Quick overview
Hallucination remains a critical challenge for developing trustworthy Multimodal Large Language Models (MLLMs). Existing benchmarks mainly focus on entity, attribute, and relation hallucinations, while knowledge-related failures are often investigated separately, lacking a unified evaluation framework. KnowHal explicitly incorporates knowledge hallucination into multimodal hallucination evaluation, spanning four dimensions: entity, attribute, relation, and knowledge. It constructs paired positive and negative questions over shared images and entities, enabling controlled comparisons among perceptual errors, knowledge-related errors, and false-premise acceptance. The benchmark contains 1,800 samples across 10 domains and 50 categories, built through a semi-automated pipeline combining LLM assistance, CLIP-based filtering, and human verification. Evaluations on 14 representative MLLMs show that the knowledge dimension consistently presents the greatest challenge for nearly all models, while most models exhibit substantial performance degradation on negative questions, revealing limited robustness to false premises. By unifying four hallucination dimensions with paired question design, KnowHal addresses an important gap and enables more comprehensive assessment of hallucinations in MLLMs.
- Introduces KnowHal, the first benchmark unifying knowledge hallucination with entity, attribute, and relation hallucination in one evaluation framework.
- Uses paired positive and negative questions over shared images and entities for controlled comparison of perceptual errors, knowledge-related errors, and false-premise acceptance.
- Contains 1,800 samples across 10 domains and 50 categories, built with LLM assistance, CLIP-based filtering, and human verification.
- Evaluates 14 MLLMs, showing knowledge dimension is the hardest challenge and negative questions significantly degrade performance.
Full text
This is an AI-generated Chinese digest; use "View original" for the source text.
Related items
85
FeaturedSetoka: A Benchmark for Hierarchical User Understanding in Personalized Agents over Heterogeneous Data
Setoka：面向异构数据个性化Agent的层级用户理解基准
This paper is well worth attention, especially for researchers and engineers working on personalized agents, dialogue systems, and memory-augmented AI. It not only provides a benchmark but defines levels of user understanding from psychology, revealing systematic failures of current systems at higher levels. If you are designing memory-based agents, this paper reminds you that memory systems need to support cross-temporal, cross-source abstract reasoning, not just retrieval.
85
FeaturedBAND: Bayesian Network Distribution Regression for High-Dimensional Distribution Estimation – Breaking the Curse
BAND：高维分布估计的贝叶斯网络回归方法——打破维度诅咒
This paper is a must-read for anyone working on high-dimensional statistics, probabilistic modeling, and generative AI. It provides a simple yet powerful theoretical answer: sparsity can break the curse of dimensionality, and Bayesian networks offer a natural framework. Compared to black-box methods like VAEs and GANs, BAND is interpretable and provably convergent, especially valuable for industry needs like reliable confidence intervals. If you research time series forecasting or risk management, this paper may change how you approach high-dimensional density estimation.
84
FeaturedCycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas
循环一致与不确定性感知的托卡马克边缘等离子体神经代理模型
This paper is worth attention because it demonstrates how advanced AI techniques (cycle consistency, uncertainty quantification) directly address a core computational bottleneck in nuclear fusion. For fusion researchers and engineers, it offers a practical tool to speed up simulations by orders of magnitude without sacrificing accuracy; for AI practitioners, it provides an excellent case of self-supervised learning and inverse problem solving in scientific applications.
83
FeaturedEquivariant Learning of a Transferable Three-Dimensional Classical Density Functional
等变学习可迁移的三维经典密度泛函
The most striking aspect of this paper is its direct coupling of machine learning to a core problem in statistical mechanics, using equivariant networks to learn a transferable density functional without any labels. For researchers in liquid physics or soft matter, this could be a new paradigm: do we need to run separate simulations for every new system anymore? For AI researchers, it also shows how to embed physical symmetries and variational principles into models rather than just fitting data. I recommend focusing on the methods, especially how equivariance and variational consistency are implemented—that's the real essence.
83
FeaturedComputational and Statistical Guarantees of the c-Rectified Flow
c-Rectified Flow 的计算与统计保证
This paper moves rectified flow from 'works in practice' to 'theoretically controllable,' which is especially important for those working on the underlying algorithms of generative models. If you are tuning pipelines like FLUX or SD3, understanding that ordinary rectified flow can fail when covariances do not commute may help you avoid hidden training pitfalls.
83
FeaturedLearning Deployable Manipulation Policies from High-Fidelity UMI Data Alone
从高保真UMI数据学习可部署操作策略
The most striking point of this paper is proving that pure robot-free data can match real teleoperation performance, breaking the convention of needing a small real-robot anchor for post-training. For research groups looking to reduce hardware costs and industrial robot application developers, HiFi-UMI's data production system and open-source dataset are directly usable resources. Pay attention to the engineering details of head-mounted SLAM and microsecond synchronization—they are key to achieving the reported accuracy.
83
FeaturedCapital Markets LLM Reliability Score: From Plausible to Bankable
资本市场LLM可靠性评分：从“看似合理”到“可银行化”
This paper is valuable for both finance and AI communities: it provides an actionable, workflow-oriented evaluation framework that's closer to real needs than generic benchmarks. If you're deploying LLMs in fintech or responsible for explainability/auditability, this is a must-read. The finding on decision usefulness is particularly striking—judges agree it's critical, yet models vary the most on it.
83
FeaturedWhere Should Optimizer State Live? Tiered State Allocation for Memory-Efficient Mixture-of-Experts Training
优化器状态应放在哪里？面向内存高效MoE训练的分层状态分配
This paper is highly valuable for teams training MoE models: it demonstrates that with careful optimizer state allocation, one can dramatically reduce memory without sacrificing accuracy—and even improve performance. It is especially relevant for labs or enterprises constrained by 40 GB or 80 GB GPU memory. The tiered strategy and ablation on momentum retention are worth studying for optimizing training frameworks like DeepSpeed or Megatron.