38 papers
CORE: Coarse-to-Fine Search with Context Management for Deep Research
Hoang Tran Vuong, Ngoc Quan Pham, Hung Bui +2
Deep research agents answer complex questions through repeated search. They rely on a harness that runs their tool calls and returns evidence to their context. This creates a trade…
RankLane: Efficient MoE Prefill on Accelerators with Ranked Expert Capacities
Wentao Ni, Chihao Yu, Hong Cai +4
Mixture-of-experts (MoE) models expand model capacity by activating only a subset of experts for each token. On accelerators with static execution plans, expert capacities and plac…
TileWeave: Heterogeneous Tiling for Sparse Prefill on Mobile SoCs
Wentao Ni, Hong Cai, Mingu Lee +7
Smartphone applications increasingly require large language models to process long contexts, whereas high prefill latency and memory demands constrain local inference. Block-sparse…
SyncVid: World-Equivariant Diffusion for Synchronized Video Generation
Farhad G. Zanjani, Animesh Karnewar, Fatih Porikli +1
Generating multiple videos of a dynamic scene from a single reference video requires synthesizing both observed and unobserved content consistently across viewpoints. Existing came…
How Much Can We Prune? Exploring the Performance Limits of Visual Token Pruning
Janghoon Cho, Shubhankar Borse, Reza Pourreza +4
Visual token pruning is increasingly used to reduce the memory and inference costs of vision-language models. Existing work has largely focused on designing method-specific selecto…
Graph Coarsening Enables Escaping Local Optima in Compute-Constrained Nonconvex MAX-CUT
M. Harper Langston, Pierre-David Letourneau, Dalton Jones +7
Because its binary optimization landscape contains exponentially many local optima, MAX-CUT is highly sensitive to initialization. We study Particle Polynomial Relaxation (PPR), a…
MagnifiQ: Patch-aware Text Guided Progressive Upscaling for High-Resolution Image Restoration
Mahesh Reddy, Yashesh Savani, Antoine Mercier +3
High-resolution image restoration from degraded inputs is challenging because it must preserve global structural consistency while recovering fine-grained local details, especially…
SNM-VFI: Symmetric Nonlinear Motion-Guided Generative Video Frame Interpolation
Jisoo Jeong, Hong Cai, Jamie Menjay Lin +5
We propose Symmetric Nonlinear Motion-guided Generative Video Frame Interpolation (SNM-VFI), a training-free framework for motion-controllable generative video frame interpolation…
MobileWan: Closing the Quality Gap for Mobile Video Diffusion
Mohsen Ghafoorian, Denis Korzhenkov, Adil Karjauv +9
Recent advances in video diffusion have been driven by scaling transformer-based architectures to billions of parameters, substantially improving visual fidelity and motion coheren…
Rethinking RAG in Long Videos: What to Retrieve and How to Use It?
Yuho Lee, Jisu Shin, Nicole Hee-Yeon Kim +5
Retrieval-augmented generation is extending beyond text to long videos, where query-relevant chunks can be represented across multiple modalities and temporal granularities. Progre…
MAPLE: Latent Multi-Agent Play for End-to-End Autonomous Driving
Rajeev Yasarla, Deepti Hegde, Hsin-Pai Cheng +9
Vision-language-action (VLA) models are effective as end-to-end motion planners, but can be brittle when evaluated in closed-loop settings due to being trained under traditional im…
CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers
Zhuojin Li, Hsin-Pai Cheng, Hong Cai +2
Diffusion Transformers (DiTs) deliver remarkable image and video generation quality but incur high computational cost, limiting scalability and on-device deployment. We introduce C…
Memory-Efficient Fine-Tuning Diffusion Transformers via Dynamic Patch Sampling and Block Skipping
Sunghyun Park, Jeongho Kim, Hyoungwoo Park +6
Diffusion Transformers (DiTs) have significantly enhanced text-to-image (T2I) generation quality, enabling high-quality personalized content creation. However, fine-tuning these mo…
ForeSea: AI Forensic Search with Multi-modal Queries for Video Surveillance
Hyojin Park, Yi Li, Janghoon Cho +8
Despite decades of work, surveillance still struggles in searching and reasoning about specific targets across long, multi-camera videos. Existing methods - tracking, retrieval, an…
A Comparative analysis of Layer-wise Representational Capacity in AR and Diffusion LLMs
Raghavv Goel, Risheek Garrepalli, Sudhanshu Agrawal +3
Autoregressive (AR) language models build representations incrementally via left-to-right prediction, while diffusion language models (dLLMs) are trained through full-sequence deno…
Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs
Wentao Ni, Kangqi Zhang, Zhongming Yu +7
As long-context inference becomes central to large language models (LLMs), attention over growing key-value caches emerges as a dominant decoding bottleneck, motivating sparse atte…
Generative Scenario Rollouts for End-to-End Autonomous Driving
Rajeev Yasarla, Deepti Hegde, Shizhong Han +11
Vision-Language-Action (VLA) models are emerging as highly effective planning models for end-to-end autonomous driving systems. However, current works mostly rely on imitation lear…
Do-Undo Bench: Reversibility for Action Understanding in Image Generation
Shweta Mahajan, Shreya Kadambi, Hoang Le +4
We introduce the Do-Undo task and benchmark to address a critical gap in vision-language models: understanding and generating plausible scene transformations driven by real-world a…
Attention Guided Alignment in Efficient Vision-Language Models
Shweta Mahajan, Hoang Le, Hyojin Park +3
Large Vision-Language Models (VLMs) rely on effective multimodal alignment between pre-trained vision encoders and Large Language Models (LLMs) to integrate visual and textual info…
Ar2Can: An Architect and an Artist Leveraging a Canvas for Multi-Human Generation
Shubhankar Borse, Phuc Pham, Farzad Farhadzadeh +6
Despite recent advances in personalized image generation, existing models consistently fail to produce reliable multi-human scenes, often merging or losing facial identity. We pres…
FLoC: Facility Location-Based Efficient Visual Token Compression for Long Video Understanding
Janghoon Cho, Jungsoo Lee, Munawar Hayat +3
Recent studies in long video understanding have harnessed the advanced visual-language reasoning capabilities of Large Multimodal Models (LMMs), driving the evolution of video-LMMs…
Resolving the Identity Crisis in Text-to-Image Generation
Shubhankar Borse, Farzad Farhadzadeh, Munawar Hayat +1
State-of-the-art text-to-image models suffer from a persistent identity crisis when generating scenes with multiple humans: producing duplicate faces, merging identities, and misco…
Generalized Contrastive Learning for Universal Multimodal Retrieval
Jungsoo Lee, Janghoon Cho, Hyojin Park +4
Despite their consistent performance improvements, cross-modal retrieval models (e.g., CLIP) show degraded performances with retrieving keys composed of fused image-text modality (…
Structuring The Future: Diffusion LLM Speculative Decoding via Calibrated Draft Graphs
Sudhanshu Agrawal, Risheek Garrepalli, Raghavv Goel +3
Diffusion LLMs (dLLMs) have recently emerged as a powerful alternative to autoregressive LLMs (AR-LLMs) with the potential to operate at significantly higher token-generation rates…