Task-Level

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 23

All topics

What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agents
Sep 30, 2026Shuyang Zhang, Jianshuo ChangLarge Language Model Decisions Large Language Model Agents
The Hong Kong Polytechnic University
About the Influence of Workflow Topology on Task Intensity Prediction through Graph Learning
Sep 30, 2026Max Otto, Haci Ismail Aslan, Joel Witzke +2Task-Level Agentic Workflow Design
Dept. Electrical Engineering and Computer Science, Technical University of Berlin, Berlin, Germany
CASS: Contribution-Aware Structured Sparsity for Model Merging
Sep 28, 2026Yan Li, Guiping Cao, Meng Xu +5Continual Model Merging Soft-Token Representations
Pengcheng Laboratory, Shenzhen, China · City University of Hong Kong, Hong Kong, China · Southern University of Science and Technology, Shenzhen, China +2
Designing and Analysing Argument Mining Pipelines: Towards a Comprehensive Assessment
Sep 22, 2026Siddharth Bhargava, Sara Tonelli, Patricia Martín-RodillaArgumentation Frameworks Natural Language Processing
Fondazione Bruno Kessler, Trento, Italy · Universidade da Coruña, A Coruña, Spain · IEGPS-CSIC, Spanish National Research Council, Santiago de Compostela, Spain
Decomposition Buys Integrity, Not Yield
Sep 15, 2026Rong HeDelegation Decomposition
Agent-Integrated Software: Interaction Contracts and Continuous Assurance
Sep 12, 2026Shengcheng Yu, Chunrong Fang, Zhenyu ChenContracts Software Engineering
Technical University of Munich, Heilbronn, Germany · State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China
EMERGE-Policy: A Robot Mind Emerges Beyond a Single Policy
Aug 30, 2026Zhirui Fang, Qingchi Yu, Ziyang Chen +18Robot Policies Embodied Agents
Tsinghua University · Nanjing University of Science and Technology · Xi’an Jiaotong University +3
When Do Task Vectors Interfere? Mapping the Validity Boundaries of Weight-Space Composition
Aug 10, 2026Chencheng Zhu, Xiaoyang Li, Taotao CaiWeight-Space Task-Level
UNSW Sydney · University of Southern Queensland
MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts
Aug 10, 2026Peiwen Li, Shiyang Zhang, Yangtian Zhang +3Model-Based Multi-Agent Systems Task-Level
Yale University, New Haven, CT, USA
Learned, Relied Upon, or Necessary? Separating Checkpoint Dependence from Task-Level Value in Sheaf GNNs
Jul 28, 2026Yi LiuSheaf Neural Networks Graph Neural Networks
School of Astronomy and Space Science, University of Science and Technology of China
CT-Merging: Consensus Directions and Task-Level Scaling for LoRA Adapter Merging
Jul 18, 2026Keumseo Ryum, Joonhyuk KangContinual Model Merging Low-Rank Adaptation Adapters
KAIST
What We Talk About When We Talk About LLM Planning: Evidence for Two Distinct Planning Abilities
Jul 13, 2026Sukai Huang, Chenyuan Zhang, Fucai Ke +4Large Language Model Planning Item Response Theory
Faculty of Information Technology, Monash University
From Search to Synthesis: Training LLMs as Zero-Shot Workflow Generators
Jun 29, 2026Gan Luo, Zihan Qin, Bin Dong +1Large Language Model Workflows Data Generation Pipelines
School of Mathematics Science, Peking University
The Speedup Paradox: Rethinking Inference Speed-Quality Trade-off in Embodied Tasks
Jun 26, 2026Yujin Wang, Junli Chen, Yixuan Li +4Embodied Decode Speedup
Tsinghua University
Post-Hoc Merging is Not Enough: Many-Shot Model Merging with Loss-Gap Balancing
Jun 15, 2026Kyungjin Im, Miru Kim, Chanin Eom +1Continual Model Merging Multi--Task Learning
Soongsil University, Republic of Korea · Department of Electrical and Computer Engineering, Sungkyunkwan University, Republic of Korea
FMplex: Model Virtualization for Serving Extensible Foundation Models
Jun 8, 2026Hetvi Shastri, Pragya Sharma, Walid A. Hanafy +3Wireless Foundation Models Large Language Model Serving
University of Massachusetts Amherst · University of California Los Angeles
Characterization of Multi-Model Agentic AI Systems on General Tasks via Trace-Driven Simulation
Jun 1, 2026Donghwan Kim, Prakhar Singh, Younghoon Min +3Production Agentic Systems Multimodal Agents
The Pennsylvania State University · SK Hynix · KAIST
PoisonForge: Task-Level Targeted Poisoning Benchmark for Instruction-Tuned LLMs
May 22, 2026Luze Sun, Anshuman Suri, Harsh Chaudhari +2Attacker Large Language Model Poisoning
Department of Computer Science Northeastern University · DatologyAI
Active Sensing Subserves Task-Level Control
May 21, 2026Andrew Lamperski, Debojyoti Biswas, Eric S. Fortune +3Viewpoint-Dependent Active Perception Active Flow Control
Department of Electrical and Computer Engineering, University of Minnesota · Laboratory for Computational Sensing and Robotics, Johns Hopkins University · 3Federated Department of Biological Sciences, New Jersey Institute of Technology +3
Interference-Aware Multi-Task Unlearning
May 18, 2026Ying-Hua Huang, Rui Fang, Hsi-Wen Chen +1Machine Unlearning Multi--Task Learning
National Taiwan University
Will AI Agents Free Us From Meaningless Work? A Human-Centered Analysis
May 15, 2026Davide Ghia, Jaspreet Ranjit, Tania Cerquitelli +1Artificial Intelligence Agents Human-Centered Framework
Politecnico di Torino · Italy · University of Southern California +3
AHASD: Asynchronous Heterogeneous Architecture for LLM Adaptive Drafting Speculative Decoding on Mobile Devices
Apr 28, 2026Ma Zirui, Fan Zhihua, Li Wenxing +4Speculative Decoding Task-Level
State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences
Hive: A Multi-Agent Infrastructure for Algorithm- and Task-Level Scaling
Apr 19, 2026Zizhang Luo, Yuhao Luo, Youwei Xiao +3Agentic Inference Task-Level
Peking University Beijing, China

Sep 30, 2026, cs.AIJ/K move · Enter open · S save

What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agents

Shuyang Zhang, Jianshuo Chang

The Hong Kong Polytechnic University

Large Language Model Decisions Large Language Model Agents Task-Level Systematic Review

Open paper AskPDF arXiv

Background. A component replacement in a language-model agent changes an execution trajectory, potentially altering later observations, resource use, and recovery opportunities. Different evidence is needed to assess its task-level benefit and the contribution of local decision quality. Methods. This critical scoping review maps 348 studies and examines 90 comparison records: 88 from 40 included studies and two from supplementary studies. Eight purposively selected cases structure the synthesis around the replaced decision, executed conditions, measurement comparability, controls, and remaining explanations. Results. Of 222 studies reporting local decision metrics, 142 also report measured task endpoints and 49 report proxies. These counts identify studies that report both types of measurement, without establishing that the measurements come from matched comparisons. Outcome Monitors reports a package-level completion gain whose attribution to detector quality remains limited; First-chunk selection reports a local improvement assessed against an offline proxy endpoint; Evidence-Carrying Termination reports fewer premature unsupported terminations and completion non-inferiority, without establishing completion superiority. Cross-case analysis identifies three candidate mechanisms involving recovery and disruption, intervention timing, and downstream use. Attribution and deployment depend on the comparison controls, label definitions, and information available to the controller. Conclusions. The review distinguishes the task-level benefit of a component replacement from the contribution of local decision quality and derives eight claim-specific reporting items. Neither online execution nor simultaneous gains in local and task metrics alone establish that better local decisions explain the task-level gain.

Momentum

Related topics

Latest papers 23