An MLLM-guided architecture with a mixture of frequency experts and relational alignment loss achieves state-of-the-art all-in-one image restoration, outperforming prior methods by up to 1.35 dB on the CDD11 dataset.
hub
In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
years
2026 12roles
background 1polarities
background 1representative citing papers
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.
AnchorPrune prunes visual tokens by first selecting a protected query-relevance anchor and then greedily adding important, non-redundant context, preserving up to 97.6% of full-token accuracy with only 160 of 2,880 tokens.
ReShift is a reasoning-level backdoor framework for VLMs that uses poisoned data construction and joint optimization to shift CoT trajectories on trigger while preserving surface coherence.
Audit of KB-VQA benchmarks reveals systematic violations of answer derivability, question clarity, and visual disambiguation assumptions, with new repair and multi-entity augmentation protocols producing different model performance trends.
ADAPT reduces MLLM hallucinations 40-60% by aligning cross-attention dynamics via visual anchors, supervised inference, and preference tuning while preserving general capabilities.
SurgCheck benchmark reveals that vision-language models for surgical VQA often depend on linguistic shortcuts rather than visual reasoning, shown by consistent performance drops on less-biased questions.
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
ReflectCAP distills model-specific hallucination and oversight patterns into Structured Reflection Notes that steer LVLMs toward more factual and complete image captions, reaching the Pareto frontier on factuality-coverage trade-offs.
MonoIR-RS synthesizes 600K infrared remote-sensing images from visible sources, rewrites captions to be IR-aware, and shows that IR-aware fine-tuning improves CLIP retrieval by up to 12.8 points and drives VLM infrared-cue coverage to 100% with near-zero RGB-color leakage.
A sandbox-trained multimodal search agent with process-oriented rewards transfers zero-shot to real Google Search and outperforms prior methods on FVQA, InfoSeek, and MMSearch.
TCA-Captioner introduces an Observer-Checker-Corrector refinement loop and TCA-Bench to address modality detachment and temporal incoherence in audiovisual video captioning.
citing papers explorer
-
Leveraging Multimodal Large Language Models for All-in-One Image Restoration via a Mixture of Frequency Experts
An MLLM-guided architecture with a mixture of frequency experts and relational alignment loss achieves state-of-the-art all-in-one image restoration, outperforming prior methods by up to 1.35 dB on the CDD11 dataset.
-
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.
-
AnchorPrune: Relevance-Anchored Contextual Expansion for Visual Token Pruning
AnchorPrune prunes visual tokens by first selecting a protected query-relevance anchor and then greedily adding important, non-redundant context, preserving up to 97.6% of full-token accuracy with only 160 of 2,880 tokens.
-
ReShift: Aha-Moment-Driven Reasoning-Level Backdoor Attacks on Vision-Language Models
ReShift is a reasoning-level backdoor framework for VLMs that uses poisoned data construction and joint optimization to shift CoT trajectories on trigger while preserving surface coherence.
-
Identifying and Resolving Pitfalls of Knowledge-Based VQA Benchmarks: Auditing, Repairing, and Augmenting
Audit of KB-VQA benchmarks reveals systematic violations of answer derivability, question clarity, and visual disambiguation assumptions, with new repair and multi-entity augmentation protocols producing different model performance trends.
-
ADAPT: Attention Dynamics Alignment with Preference Tuning for Faithful MLLMs
ADAPT reduces MLLM hallucinations 40-60% by aligning cross-attention dynamics via visual anchors, supervised inference, and preference tuning while preserving general capabilities.
-
SurgCheck: Do Vision-Language Models Really Look at Images in Surgical VQA?
SurgCheck benchmark reveals that vision-language models for surgical VQA often depend on linguistic shortcuts rather than visual reasoning, shown by consistent performance drops on less-biased questions.
-
V-tableR1: Process-Supervised Multimodal Table Reasoning with Critic-Guided Policy Optimization
V-tableR1 uses a critic VLM for dense step-level feedback and a new PGPO algorithm to shift multimodal table reasoning from pattern matching to verifiable logical steps, achieving SOTA accuracy with a 4B open-source model.
-
ReflectCAP: Detailed Image Captioning with Reflective Memory
ReflectCAP distills model-specific hallucination and oversight patterns into Structured Reflection Notes that steer LVLMs toward more factual and complete image captions, reaching the Pareto frontier on factuality-coverage trade-offs.
-
MonoIR-RS: Infrared Remote Sensing Vision-Language Learning with CLIP and VLM Adaptation
MonoIR-RS synthesizes 600K infrared remote-sensing images from visible sources, rewrites captions to be IR-aware, and shows that IR-aware fine-tuning improves CLIP retrieval by up to 12.8 points and drives VLM infrared-cue coverage to 100% with near-zero RGB-color leakage.
-
ProMMSearchAgent: A Generalizable Multimodal Search Agent Trained with Process-Oriented Rewards
A sandbox-trained multimodal search agent with process-oriented rewards transfers zero-shot to real Google Search and outperforms prior methods on FVQA, InfoSeek, and MMSearch.
-
Temporal and Cross-Modal Alignment for Enhanced Audiovisual Video Captioning
TCA-Captioner introduces an Observer-Checker-Corrector refinement loop and TCA-Bench to address modality detachment and temporal incoherence in audiovisual video captioning.