Task Interference

Follow this topic Ask about this topic

Latest papers

12 papers2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

FlexUMM: Flexible Representation Learning for Unified Multimodal Models
Oct 15, 2026Yinyi Luo, Haoyue Bai, Yi Ge +3Task Interference Unified Multimodal Models
Carnegie Mellon University · William & Mary · University of Maryland
The Bell Boundary of Shared Representations
Oct 12, 2026Yang Wu, Hanyu Zhao, Yazhou Zhang +1Shared Representation Learning Task Interference
Tianjin University, China
Passes Alone, Fails Together: Benchmarking Semantic Coordination in Parallel LLM-Agent Development
Sep 21, 2026Haocheng Xia, Eugene Wu, Yongjoo ParkLLM Agent Evaluation AI Agent Benchmarks
University of Illinois Urbana-Champaign Champaign, USA · Columbia University New York, USA
When Muon Meets Task Interference: A Spectral Perspective on Continual Learning and Model Merging
Aug 27, 2026Shangge Liu, Yuehan Yin, Yinghuan Shi +2Task Interference Continual Learning
State Key Laboratory for Novel Software Technology, Nanjing University · University of Wollongong
Reading the Gate, Not the Interference: Output-Side Interference Measurement Does Not Track Merge Collapse
Aug 12, 2026Chencheng ZhuTask Vector Merging Task Arithmetic
UNSW Sydney
Interference and Retention in Continual Learning
Jul 10, 2026Julius StörkContinual Learning Online Continual Learning
VARTA Microbattery GmbH, Ellwangen, Germany
Post-Hoc Merging is Not Enough: Many-Shot Model Merging with Loss-Gap Balancing
Jun 15, 2026Kyungjin Im, Miru Kim, Chanin Eom +1Multi-Task Learning Language Model Merging
Soongsil University, Republic of Korea · Department of Electrical and Computer Engineering, Sungkyunkwan University, Republic of Korea
Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retention
May 28, 2026Jing Huang, Daniel Wurgaft, Rachit Bansal +6Long-Tail Learning Language Model Scaling Laws
Stanford University · Kempner Institute at Harvard University · MIT +1
Crowded in B-Space: Calibrating Shared Directions for LoRA Merging
Apr 18, 2026Yixuan Tang, Yi YangAdapter Tuning Low-Rank Adaptation
The Hong Kong University of Science and Technology
Did You Forget What I Asked? Prospective Memory Failures in Large Language Models
Mar 7, 2026Avni MittalInstruction Following LLM Reliability
Model Merging in the Essential Subspace
Feb 23, 2026Longhua Li, Lei Qi, Qi Tian +1Task Interference Model Merging
School of Computer Science and Engineering, Southeast University, Nanjing, China · Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications (Southeast University), Ministry of Education, China · Huawei Technologies, Shanghai, China
Mitigating Task Objective Conflicts for Continual Learning in Unified Multimodal Models
Jun 4, 2025Jiaxing Zhang, Hao TangTask Interference Unified Multimodal Models
Sichuan University, Chengdu, China · Peking University, Beijing, China

Oct 15, 2026J/K move · Enter open · S save

FlexUMM: Flexible Representation Learning for Unified Multimodal Models

Yinyi Luo, Haoyue Bai, Yi Ge, Nilton Cesar Rojas Vales+2

Carnegie Mellon University · William & Mary · University of Maryland

Task Interference Unified Multimodal Models Multimodal Learning Conditional Computation

Open paper AskPDF arXiv

Unified multimodal models (UMMs) can integrate understanding and generation within a single backbone. However, it is commonly known that these two capabilities can cause conflicting representations, leading to a central question: how representation and computation should be shared across functionalities? Existing UMMs often rely on a largely fixed structure, where different inputs and functionalities follow the same depth and parameter-sharing pattern. We argue that this assumption is too restrictive: more effective unification requires sharing the computation that supports cross-functional communication while specializing the capacity that becomes task- or input-specific. We introduce FlexUMM, a flexible UMM framework that dynamically decides when to preserve shared computation, when to route specialized capacity, and when to stop further processing. Evidence shows that this flexible allocation improves UMMs along two complementary axes. It improves accuracy by reducing harmful competition between functionalities, and it improves efficiency by avoiding unnecessary computation once an input representation is sufficiently mature. Beyond simple architectural unification from the pre-trained models, FlexUMM works as a model-agnostic, functional plug-in to make existing UMMs more unified, stronger, and efficient. Code is available at: https://github.com/AIFrontierLab/TorchUMM/tree/main/src/umm/post_training/flexumm.