VLMs hallucinate by prioritizing contradictory on-screen text over visual content, addressed via the VisualTextTrap benchmark with 6,057 human-validated samples and the VTHM-MoE dual-encoder framework using dimension-specific experts and adaptive routing.
Canonical reference
Videohallucer: Evaluating intrinsic and extrinsic hallucinations in large video-language models
Canonical reference. 80% of citing Pith papers cite this work as background.
6
Pith papers citing it
Background
80% of classified citations
citation-role summary
background 4
dataset 1
citation-polarity summary
years
2026 6representative citing papers
TOC-Bench is a new diagnostic benchmark that reveals major weaknesses in temporal object consistency for Video-LLMs, including event counting, ordering, identity reasoning, and hallucination avoidance.
Video-LLMs fail physical reasoning due to semantic prior dominance rather than perception deficits; a new programmatic adversarial curriculum and visual-anchored reasoning chain enable substantial gains via standard LoRA fine-tuning.
Video-ToC adds tree-guided cue localization, demand-based RL rewards, and automated datasets to video LLMs, reporting better results than prior methods on six understanding benchmarks plus a hallucination test.
Raven automates Scratch program assessment by having instructors specify task-level video generation rules and using LLMs to analyze resulting videos for behavioral compliance, outperforming prior tools on real student submissions.
Decoder-side Temporal Rebalancing (DTR) reduces hallucinations in Video-LLMs by mitigating over-dominance of a single anchor frame during inference without training or auxiliary models.
citing papers explorer
-
When Text Hijacks Vision: Benchmarking and Mitigating Text Overlay-Induced Hallucination in Vision Language Models
VLMs hallucinate by prioritizing contradictory on-screen text over visual content, addressed via the VisualTextTrap benchmark with 6,057 human-validated samples and the VTHM-MoE dual-encoder framework using dimension-specific experts and adaptive routing.
-
TOC-Bench: A Temporal Object Consistency Benchmark for Video Large Language Models
TOC-Bench is a new diagnostic benchmark that reveals major weaknesses in temporal object consistency for Video-LLMs, including event counting, ordering, identity reasoning, and hallucination avoidance.
-
From Priors to Perception: Grounding Video-LLMs in Physical Reality
Video-LLMs fail physical reasoning due to semantic prior dominance rather than perception deficits; a new programmatic adversarial curriculum and visual-anchored reasoning chain enable substantial gains via standard LoRA fine-tuning.
-
Video-ToC: Video Tree-of-Cue Reasoning
Video-ToC adds tree-guided cue localization, demand-based RL rewards, and automated datasets to video LLMs, reporting better results than prior methods on six understanding benchmarks plus a hallucination test.
-
Raven: Rethinking Automated Assessment for Scratch Programs via Video-Grounded Evaluation
Raven automates Scratch program assessment by having instructors specify task-level video generation rules and using LLMs to analyze resulting videos for behavioral compliance, outperforming prior tools on real student submissions.
-
Relaxing Anchor-Frame Dominance for Mitigating Hallucinations in Video Large Language Models
Decoder-side Temporal Rebalancing (DTR) reduces hallucinations in Video-LLMs by mitigating over-dominance of a single anchor frame during inference without training or auxiliary models.