AVID is the first large-scale benchmark for audio-visual inconsistency detection, grounding, classification, and reasoning in long videos, constructed via agent-driven methods and showing that state-of-the-art models struggle while a fine-tuned baseline improves performance.
hub Mixed citations
AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation
Mixed citation behavior. Most common role is background (67%).
35
Pith papers citing it
Background
67% of classified citations
abstract
Despite making significant progress in multi-modal tasks, current Multi-modal Large Language Models (MLLMs) encounter the significant challenge of hallucinations, which may lead to harmful consequences. Therefore, evaluating MLLMs' hallucinations is becoming increasingly important in model improvement and practical application deployment. Previous works are limited in high evaluation costs (e.g., relying on humans or advanced LLMs) and insufficient evaluation dimensions (e.g., types of tasks and hallucinations). In this paper, we propose an LLM-free multi-dimensional benchmark AMBER, which can be used to evaluate both generative task and discriminative task including existence, attribute and relation hallucination. Based on AMBER, we design a low-cost and efficient evaluation pipeline. Additionally, we conduct a comprehensive evaluation and detailed analysis of mainstream MLLMs including GPT-4V(ision), and also give guideline suggestions for mitigating hallucinations. The data and code of AMBER are available at https://github.com/junyangwang0410/AMBER.
hub tools
citation-role summary
background 4
baseline 1
dataset 1
citation-polarity summary
representative citing papers
MM-Snowball benchmark diagnoses hallucination snowballing in multi-turn MLLM dialogues; CAVR mitigates it via dual visual rectification at representation and logit levels.
YARD is a training-free method using Y-shaped decoder architecture and register tokens to improve contrastive decoding for hallucination reduction in LVLMs with lower latency.
SIRA mitigates hallucinations in LVLMs by internally contrasting full visual access against a masked late-layer branch that retains shared context but lacks fine-grained visual evidence.
OxyEcomBench is a unified multimodal benchmark covering 6 capability areas and 29 tasks with authentic e-commerce data to measure how well foundation models handle real platform, merchant, and customer challenges.
DO-Bench is a controlled benchmark that attributes VLM object hallucination errors to textual prior pressure, perceptual limits, or their interaction via two diagnostic dimensions and metrics.
ZINA detects fine-grained hallucinations in MLLM outputs, classifies errors into six types, and proposes edits, outperforming GPT-4o and Llama-3.2 on the new VisionHall dataset of annotated and synthetic samples.
VidHal is a new benchmark that evaluates VLLM temporal hallucinations through a caption ordering task on videos with varying hallucination levels.
IC-VCO places contrastive images in one context for a consistent DPO-style objective, adds Visual Contrast Distillation, and uses semantic perturbation for hard negatives, reporting best results on five benchmarks.
Deep Pre-Alignment uses a small VLM perceiver instead of ViT to pre-align visual features with LLM text space, yielding 1.9-3.0 point gains on multimodal benchmarks and 32.9% less language forgetting.
Vision-language models contain identifiable grounding and hallucination pathways; suppressing the latter reduces object hallucinations by up to 76% while preserving accuracy.
Gaze Attention groups visual embeddings into selectable regions and dynamically restricts attention to task-relevant ones, matching dense baselines with up to 90% fewer visual KV entries via added context tokens.
LVLMs show vocabulary hijacking by inert tokens that decode to hijacking anchors; HABI locates them, NHAR finds resilient heads, and HAVAE boosts those heads to cut hallucinations.
CAVI framework uses character-guided token pruning, orthogonal feature modulation, and modality-adaptive role steering to resolve modality-role interference in multimodal RPAs.
R-CoV is a six-step region-aware chain-of-verification technique that elicits coordinate and description outputs from LVLMs themselves to detect and reduce object hallucinations without external models or retraining.
PSRD mitigates visual hallucinations in LVLMs via phase-wise self-reward decoding, cutting rates by 50% on LLaVA-1.5-7B and outperforming prior methods on five benchmarks.
SIF creates semantically in-distribution fingerprints for LVLMs by distilling text watermarks into visual inputs and optimizing for robustness against detection and modification.
VGA constructs precise visual grounding from token semantics to guide MLLM attention toward relevant regions, dynamically suppressing described areas in captioning, and achieves SOTA dehallucination with negligible overhead.
ORCA is an agentic reasoning framework that enhances factual accuracy and adversarial robustness of pretrained LVLMs via an Observe-Reason-Critique-Act loop with small vision models, reporting accuracy gains of up to 40% on hallucination benchmarks and 20% under adversarial perturbations.
TARS uses token-adaptive min-max preference optimization and FFT-based spectral regularization to cut hallucination rates in MLLMs from 26.4% to 13.2% with only 4.8k samples, outperforming standard DPO and larger data-augmented baselines.
SENTINEL reduces MLLM object hallucinations by over 90% via sentence-level early intervention with detector-bootstrapped preference data and C-DPO loss, outperforming prior SOTA on hallucination and capability benchmarks.
CHASD is an inference-time framework that gates contrastive decoding via an uncertainty threshold and constructs negative branches through attention-guided perturbations of salient visual tokens to mitigate hallucinations in LVLMs.
UE-DPO quantifies epistemic uncertainty from grounding failures to direct more learning pressure on hard visual tokens in preferred samples while easing penalties on dispreferred ones.
MPD reduces hallucinations in LVLMs by 23.4% while retaining 97.4% of general capability through semantic disentanglement and selective parameter updates.
citing papers explorer
-
AVID: A Benchmark for Omni-Modal Audio-Visual Inconsistency Understanding via Agent-Driven Construction
AVID is the first large-scale benchmark for audio-visual inconsistency detection, grounding, classification, and reasoning in long videos, constructed via agent-driven methods and showing that state-of-the-art models struggle while a fine-tuned baseline improves performance.
-
MM-Snowball: Evaluating and Mitigating Hallucination Snowballing in Multimodal Multi-Turn Dialogue
MM-Snowball benchmark diagnoses hallucination snowballing in multi-turn MLLM dialogues; CAVR mitigates it via dual visual rectification at representation and logit levels.
-
YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models
YARD is a training-free method using Y-shaped decoder architecture and register tokens to improve contrastive decoding for hallucination reduction in LVLMs with lower latency.
-
Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution
SIRA mitigates hallucinations in LVLMs by internally contrasting full visual access against a masked late-layer branch that retains shared context but lacks fine-grained visual evidence.
-
OxyEcomBench: Benchmarking Multimodal Foundation Models across E-Commerce Ecosystems
OxyEcomBench is a unified multimodal benchmark covering 6 capability areas and 29 tasks with authentic e-commerce data to measure how well foundation models handle real platform, merchant, and customer challenges.
-
DO-Bench: An Attributable Benchmark for Diagnosing Object Hallucination in Vision-Language Models
DO-Bench is a controlled benchmark that attributes VLM object hallucination errors to textual prior pressure, perceptual limits, or their interaction via two diagnostic dimensions and metrics.
-
ZINA: Multimodal Fine-grained Hallucination Detection and Editing
ZINA detects fine-grained hallucinations in MLLM outputs, classifies errors into six types, and proposes edits, outperforming GPT-4o and Llama-3.2 on the new VisionHall dataset of annotated and synthetic samples.
-
VidHal: Benchmarking Temporal Hallucinations in Vision LLMs
VidHal is a new benchmark that evaluates VLLM temporal hallucinations through a caption ordering task on videos with varying hallucination levels.
-
Learning from Fine-Grained Visual Discrepancies: Mitigating Multimodal Hallucinations via In-Context Visual Contrastive Optimization
IC-VCO places contrastive images in one context for a consistent DPO-style objective, adds Visual Contrast Distillation, and uses semantic perturbation for hard negatives, reporting best results on five benchmarks.
-
Deep Pre-Alignment for VLMs
Deep Pre-Alignment uses a small VLM perceiver instead of ViT to pre-align visual features with LLM text space, yielding 1.9-3.0 point gains on multimodal benchmarks and 32.9% less language forgetting.
-
Dual-Pathway Circuits of Object Hallucination in Vision-Language Models
Vision-language models contain identifiable grounding and hallucination pathways; suppressing the latter reduces object hallucinations by up to 76% while preserving accuracy.
-
Learning to See What You Need: Gaze Attention for Multimodal Large Language Models
Gaze Attention groups visual embeddings into selectable regions and dynamically restricts attention to task-relevant ones, matching dense baselines with up to 90% fewer visual KV entries via added context tokens.
-
Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination
LVLMs show vocabulary hijacking by inert tokens that decode to hijacking anchors; HABI locates them, NHAR finds resilient heads, and HAVAE boosts those heads to cut hallucinations.
-
Through the Lens of Character: Resolving Modality-Role Interference in Multimodal Role-Playing Agent
CAVI framework uses character-guided token pruning, orthogonal feature modulation, and modality-adaptive role steering to resolve modality-role interference in multimodal RPAs.
-
R-CoV: Region-Aware Chain-of-Verification for Alleviating Object Hallucinations in LVLMs
R-CoV is a six-step region-aware chain-of-verification technique that elicits coordinate and description outputs from LVLMs themselves to detect and reduce object hallucinations without external models or retraining.
-
Mitigating Multimodal Hallucination via Phase-wise Self-reward
PSRD mitigates visual hallucinations in LVLMs via phase-wise self-reward decoding, cutting rates by 50% on LLaVA-1.5-7B and outperforming prior methods on five benchmarks.
-
SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models
SIF creates semantically in-distribution fingerprints for LVLMs by distilling text watermarks into visual inputs and optimizing for robustness against detection and modification.
-
Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention
VGA constructs precise visual grounding from token semantics to guide MLLM attention toward relevant regions, dynamically suppressing described areas in captioning, and achieves SOTA dehallucination with negligible overhead.
-
ORCA: An Agentic Reasoning Framework for Hallucination and Adversarial Robustness in Vision-Language Models
ORCA is an agentic reasoning framework that enhances factual accuracy and adversarial robustness of pretrained LVLMs via an Observe-Reason-Critique-Act loop with small vision models, reporting accuracy gains of up to 40% on hallucination benchmarks and 20% under adversarial perturbations.
-
TARS: MinMax Token-Adaptive Preference Strategy for Hallucination Reduction in MLLMs
TARS uses token-adaptive min-max preference optimization and FFT-based spectral regularization to cut hallucination rates in MLLMs from 26.4% to 13.2% with only 4.8k samples, outperforming standard DPO and larger data-augmented baselines.
-
Mitigating Object Hallucinations via Sentence-Level Early Intervention
SENTINEL reduces MLLM object hallucinations by over 90% via sentence-level early intervention with detector-bootstrapped preference data and C-DPO loss, outperforming prior SOTA on hallucination and capability benchmarks.
-
CHASD: Language Increment-Calibrated Contrastive Decoding against Hallucination in LVLMs
CHASD is an inference-time framework that gates contrastive decoding via an uncertainty threshold and constructs negative branches through attention-guided perturbations of salient visual tokens to mitigate hallucinations in LVLMs.
-
Uncertainty-Aware Exploratory Direct Preference Optimization for Multimodal Large Language Models
UE-DPO quantifies epistemic uncertainty from grounding failures to direct more learning pressure on hard visual tokens in preferred samples while easing penalties on dispreferred ones.
-
Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation
MPD reduces hallucinations in LVLMs by 23.4% while retaining 97.4% of general capability through semantic disentanglement and selective parameter updates.
-
Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction
MESA reduces hallucinations in LVLMs via controlled selective latent intervention that preserves the original token distribution.
-
Steering the Verifiability of Multimodal AI Hallucinations
Researchers create a human-labeled dataset of obvious and elusive multimodal hallucinations and use learned activation-space probes to control their verifiability in MLLMs.
-
NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation
NoisyGRPO is an RL framework that perturbs visual inputs with Gaussian noise for exploration and computes trajectory advantages via Bayesian posterior fusion of noise prior and reward likelihood to improve multimodal CoT generalization.
-
Revisit What You See: Revealing Visual Semantics in Vision Tokens to Guide LVLM Decoding
ReVisiT refines LVLM output distributions during decoding by projecting selected vision tokens into text space via context-aware constrained divergence minimization.
-
Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration
CAAC mitigates hallucinations in LVLMs via Visual-Token Calibration and Adaptive Attention Re-Scaling guided by model confidence, showing gains on CHAIR, AMBER, and POPE especially in long-form generation.
-
Hallucination of Multimodal Large Language Models: A Survey
The survey organizes causes of hallucinations in MLLMs, reviews evaluation benchmarks and metrics, and outlines mitigation approaches plus open questions.
-
A Survey on Hallucination in Large Vision-Language Models
This survey reviews the definition, symptoms, evaluation benchmarks, root causes, and mitigation methods for hallucinations in large vision-language models.
-
A Survey on Multimodal Large Language Models
This survey organizes the architectures, training strategies, data, evaluation methods, extensions, and challenges of Multimodal Large Language Models.
- Do Vision-Language Models Understand 3D Scenes or Just Catalogue Objects?
- How do Humans Process AI-generated Hallucination Contents: a Neuroimaging Study
- MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs