Introduces VidPair-Halluc benchmark of 1K background-controlled adversarial video pairs and 11K QA pairs generated via PairFlow pipeline to evaluate hallucination in LVMs.
arXiv preprint arXiv:2504.13169 , year=
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 9roles
background 2polarities
background 2representative citing papers
CAS mitigates object hallucinations in MLLMs by extracting two context preference vectors from designed conflict samples and applying signed residual injection at mid-early MLP layers without retraining or added latency.
Audio-Contrastive Preference Optimization (ACPO) mitigates audio hallucination in AVLMs via output-contrastive and input-contrastive objectives that enforce faithful audio grounding.
DeP mitigates MLLM hallucinations by dynamically perturbing text prompts to identify and reinforce stable visual evidence regions while counteracting language prior biases using attention variance and logit statistics.
Phi-Nav generates path-level hindsight instructions from on-policy exploration trajectories to supply additional semantic supervision for vision-language navigation agents.
In LVLMs, attention can be replaced by random Gaussian weights with little or no performance loss, indicating that current models get lost in attention rather than efficiently using visual context.
ESC uses emotional cues triggered by an external verifier to enable training-free self-correction in VLMs, improving reliability on safety, hallucination, and reasoning benchmarks.
FineGen uses a VLM multi-agent pipeline to build FineGen-100K, a 147k-sample hierarchical dataset of attribute-specific hard negatives, reporting 96.7% validity and +14.4% downstream accuracy gain on hard samples in FG-OVD.
A 22M-parameter hyperbolic model answers structured EHR questions with accuracy close to LLM-based systems (EHRXQA 89.5%, MIMIC-Instr 76.0%).
citing papers explorer
-
No Place to Hide: Benchmarking Video Hallucination with Background-Controlled Pairs
Introduces VidPair-Halluc benchmark of 1K background-controlled adversarial video pairs and 11K QA pairs generated via PairFlow pipeline to evaluate hallucination in LVMs.
-
Rethinking Visual Neglect: Steering via Context-Preference for MLLM Hallucination Mitigation
CAS mitigates object hallucinations in MLLMs by extracting two context preference vectors from designed conflict samples and applying signed residual injection at mid-early MLP layers without retraining or added latency.
-
Don't Let the Video Speak: Audio-Contrastive Preference Optimization for Audio-Visual Language Models
Audio-Contrastive Preference Optimization (ACPO) mitigates audio hallucination in AVLMs via output-contrastive and input-contrastive objectives that enforce faithful audio grounding.
-
Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation
DeP mitigates MLLM hallucinations by dynamically perturbing text prompts to identify and reinforce stable visual evidence regions while counteracting language prior biases using attention variance and logit statistics.
-
Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation
Phi-Nav generates path-level hindsight instructions from on-policy exploration trajectories to supply additional semantic supervision for vision-language navigation agents.
-
Large Vision-Language Models Get Lost in Attention
In LVLMs, attention can be replaced by random Gaussian weights with little or no performance loss, indicating that current models get lost in attention rather than efficiently using visual context.
-
ESC: Emotional Self-Correction for Reliable Vision-Language Models
ESC uses emotional cues triggered by an external verifier to enable training-free self-correction in VLMs, improving reliability on safety, hallucination, and reasoning benchmarks.
-
FineGen: A VLM-based Multi-Agent Framework for Fine-Grained Image-Text Dataset Construction
FineGen uses a VLM multi-agent pipeline to build FineGen-100K, a 147k-sample hierarchical dataset of attribute-specific hard negatives, reporting 96.7% validity and +14.4% downstream accuracy gain on hard samples in FG-OVD.
-
HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering
A 22M-parameter hyperbolic model answers structured EHR questions with accuracy close to LLM-based systems (EHRXQA 89.5%, MIMIC-Instr 76.0%).