NeuroFlow is the first unified flow model for bidirectional visual encoding and decoding from neural activity using NeuroVAE and cross-modal flow matching.
Learning transferable visual models from natural language supervi- sion
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
A weakly supervised optimal-transport method infers frame-phrase alignments in human motion from clip-level text, evaluated on only 30 test pairs.
O2MAG generates high-fidelity text-guided anomalies from a single image without training by manipulating self-attention in diffusion models with anomaly masks and dual enhancements.
EVRB is a three-part inference-time method that prunes ambiguous visual tokens, divides the model's output distribution by a text-only prior, and triggers early stopping to reduce hallucination in LVLMs.
OmniME integrates retrospective feature supervision, motion preservation, and triplet semantic alignment to achieve state-of-the-art text-motion editing alignment on MotionFix and STANCE datasets.
citing papers explorer
-
NeuroFlow: Toward Unified Visual Encoding and Decoding from Neural Activity
NeuroFlow is the first unified flow model for bidirectional visual encoding and decoding from neural activity using NeuroVAE and cross-modal flow matching.
-
FineMoLA: Towards Fine-Grained Motion-Language Alignment from Clip-Level Supervision
A weakly supervised optimal-transport method infers frame-phrase alignments in human motion from clip-level text, evaluated on only 30 test pairs.
-
One-to-More: High-Fidelity Training-Free Anomaly Generation with Attention Control
O2MAG generates high-fidelity text-guided anomalies from a single image without training by manipulating self-attention in diffusion models with anomaly masks and dual enhancements.
-
Enhancing Visual Reliance in Text Generation: A Bayesian Perspective on Mitigating Hallucination in Large Vision-Language Models
EVRB is a three-part inference-time method that prunes ambiguous visual tokens, divides the model's output distribution by a text-only prior, and triggers early stopping to reduce hallucination in LVLMs.
-
Omni-Supervised Motion Editing: Balancing Change and Invariance through Positive-Negative Learning
OmniME integrates retrospective feature supervision, motion preservation, and triplet semantic alignment to achieve state-of-the-art text-motion editing alignment on MotionFix and STANCE datasets.