Cover Image

DeepInsert: Early Layer Bypass for Efficient and Performant Multimodal Understanding

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Choraria, Moulik, Wu, Xinbo, Bhimaraju, Akhil, Sekhar, Nitesh, Wu, Yue, Zhang, Xu, Singhal, Prateek, Varshney, Lav R.
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Semantically Grounded QFormer for Efficient Vision Language Understanding
by: Choraria, Moulik, et al.
Published: (2023)

Skip-It? Theoretical Conditions for Layer Skipping in Vision-Language Models
by: Hartman, Max, et al.
Published: (2025)

Watermarking Discrete Diffusion Language Models
by: Bagchi, Avi, et al.
Published: (2025)

Dynamic Batching of Online Arrivals to Leverage Economies of Scale
by: Bhimaraju, Akhil, et al.
Published: (2023)

Context-Gated Associative Retrieval: From Theory to Transformers
by: Choraria, Moulik, et al.
Published: (2026)

Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models
by: Hartman, Max, et al.
Published: (2026)

Transformer-based Causal Language Models Perform Clustering
by: Wu, Xinbo, et al.
Published: (2024)

A Meta-Learning Perspective on Transformers for Causal Language Modeling
by: Wu, Xinbo, et al.
Published: (2023)

Fractional Budget Allocation for Influence Maximization under General Marketing Strategies
by: Bhimaraju, Akhil, et al.
Published: (2024)

Concealment of Intent: A Game-Theoretic Analysis
by: Wu, Xinbo, et al.
Published: (2025)

Verifier Threshold: An Efficient Test-Time Scaling Approach for Image Generation
by: Sundaresha, Vignesh, et al.
Published: (2025)

A Theoretical Game of Attacks via Compositional Skills
by: Wu, Xinbo, et al.
Published: (2026)

SwitchCIT: Switching for Continual Instruction Tuning
by: Wu, Xinbo, et al.
Published: (2024)

Convex Distillation: Efficient Compression of Deep Networks via Convex Optimization
by: Varshney, Prateek, et al.
Published: (2024)

Hallucination Basins: A Dynamic Framework for Understanding and Controlling LLM Hallucinations
by: Cherukuri, Kalyan, et al.
Published: (2026)

Fed-SB: A Silver Bullet for Extreme Communication Efficiency and Performance in (Private) Federated LoRA Fine-Tuning
by: Singhal, Raghav, et al.
Published: (2025)

Learning from One and Only One Shot
by: Yu, Haizi, et al.
Published: (2022)

SparseJEPA: Sparse Representation Learning of Joint Embedding Predictive Architectures
by: Hartman, Max, et al.
Published: (2025)

Adaptively Bypassing Vision Transformer Blocks for Efficient Visual Tracking
by: Yang, Xiangyang, et al.
Published: (2024)

Efficient Model-Agnostic Multi-Group Equivariant Networks
by: Baltaji, Razan, et al.
Published: (2023)

Do Multimodal Large Language Models Understand Welding?
by: Khvatskii, Grigorii, et al.
Published: (2025)

Task-specific regularization loss towards model calibration for reliable lung cancer detection
by: Kalra, Mehar Prateek, et al.
Published: (2024)

NeRF-Insert: 3D Local Editing with Multimodal Control Signals
by: Sabat, Benet Oriol, et al.
Published: (2024)

StableIdentity: Inserting Anybody into Anywhere at First Sight
by: Wang, Qinghe, et al.
Published: (2024)

MMFuser: Multimodal Multi-Layer Feature Fuser for Fine-Grained Vision-Language Understanding
by: Cao, Yue, et al.
Published: (2024)

InverseMeetInsert: Robust Real Image Editing via Geometric Accumulation Inversion in Guided Diffusion Models
by: Zheng, Yan, et al.
Published: (2024)

Indian Sign Language Detection for Real-Time Translation using Machine Learning
by: Singhal, Rajat, et al.
Published: (2025)

Semantic Compression with Information Lattice Learning
by: Yu, Haizi, et al.
Published: (2024)

Containment Verification: AI Safety Guarantees Independent of Alignment
by: Moon, Royce, et al.
Published: (2026)

Online Reinforcement Learning with Passive Memory
by: Pattanaik, Anay, et al.
Published: (2024)

SwiftVLM: Efficient Vision-Language Model Inference via Cross-Layer Token Bypass
by: Qian, Chen, et al.
Published: (2026)

PIXELS: Progressive Image Xemplar-based Editing with Latent Surgery
by: Biswas, Shristi Das, et al.
Published: (2025)

Harmonizing Visual Representations for Unified Multimodal Understanding and Generation
by: Wu, Size, et al.
Published: (2025)

ChartSketcher: Reasoning with Multimodal Feedback and Reflection for Chart Understanding
by: Huang, Muye, et al.
Published: (2025)

Bypassing LLM Watermarks with Color-Aware Substitutions
by: Wu, Qilong, et al.
Published: (2024)

VideoAgent: A Memory-augmented Multimodal Agent for Video Understanding
by: Fan, Yue, et al.
Published: (2024)

Impact of Layer Norm on Memorization and Generalization in Transformers
by: Singhal, Rishi, et al.
Published: (2025)

Quick Bypass Mechanism of Zero-Shot Diffusion-Based Image Restoration
by: Tai, Yu-Shan, et al.
Published: (2025)

An Information Theory of Compute-Optimal Size Scaling, Emergence, and Plateaus in Language Models
by: Nayak, Anuj K., et al.
Published: (2024)

Structure-Guided Histopathology Synthesis via Dual-LoRA Diffusion
by: Xu, Xuan, et al.
Published: (2026)