Intermediate States

Follow this topic Ask about this topic

Momentum

5 papers in the last four weeks, against 1 the four weeks before. 0.1% of all new papers.

Jul 6Week of Sep 21

Latest papers 18

All topics

Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers
Sep 30, 2026Jiaheng Chen, Jiaxing Li, Yucheng Xiao +4Redundancy Scaling
Harbin Institute of Technology, Shenzhen, China · Northeastern University, Shenyang, China · Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China +2
Not Too Hard, Not Too Easy: Learning from Intermediate States for LLM Structured Reasoning
Sep 27, 2026Hongbo Chen, Guohua Lu, Ting Dang +1LLM Reasoning Strategies Intermediate States
University of Auckland, Auckland, New Zealand · University of Melbourne, Melbourne, Australia
RoboRecover: Benchmarking Robot Policy Recovery under Execution Deviations
Sep 24, 2026Yang Li, Chen Zhao, Zhuoran Wang +5Robot Policies Robot Systems
School of Information, Renmin University of China, Beijing, China · Key Laboratory of Data Engineering and Knowledge Engineering, Beijing, China · University of Science and Technology of China +1
Response-state Learning for Transferable Vibrational Spectroscopic Characterization with Electron Prior
Sep 24, 2026Zetong Li, Zhuosong Xie, Hengyu Fan +6Laser-Induced Breakdown Spectroscopy Quantum Chemistry
Innovation Center for AI and Drug Discovery, School of Pharmacy, East China Normal University, Shanghai 200062, China.
Windowed A-K-MDP
Sep 12, 2026Xiangwen Yang, Frankie Cho, Iadine ChadesMarkov Decision Processes Intermediate States
Monash University Melbourne, Australia
Staying on the Attack Path: Structured State for Long-Horizon Automated Penetration Testing
Sep 7, 2026Weizhe Wang, Yitong Zhang, Yao Zhang +4Penetration Testing Diverse Attacks
Tianjin University, Tianjin, 300072, China · Changchun University of Science and Technology, Changchun, 130022, China · Jilin Province Key Laboratory of Network and Information Security, Changchun, 130022, China +1
Persistent Semantic Entities in Tool-Augmented LLM Systems
Aug 8, 2026Zhaohui WangEntities Intermediate States
USC Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA.
Task Decomposition-Guided Reranking for Adaptive Agent Skill Retrieval
Jul 7, 2026Yanping Chen, Weijie Shi, Wen Yang +1Reranking Intermediate States
School of Computer Science and Technology, Soochow University · The Hong Kong University of Science and Technology
LedgerAgent: Structured State for Policy-Adherent Tool-Calling Agents
Jun 18, 2026Md Nayem Uddin, Amir Saeidi, Eduardo Blanco +1Ledger State-Tracking
♠Arizona State University · ♢University of Arizona
Through the PRISM: Preference Representation in Intermediate States of Video Diffusion Models
Jun 18, 2026Haoxuan Wu, Lai Man Po, Mengyang Liu +3Video Diffusion Models Video Foundation Models
City University of Hong Kong · Video Rebirth · The Chinese University of Hong Kong
SVoT: State-aware Visualization-of-Thought for Spatial Reasoning via Reinforcement Learning
Jun 10, 2026Chao Lei, Yanbei Jiang, Markus Hiller +4Spatial Reasoning Intermediate States
School of Computing and Information Systems The University of Melbourne Melbourne, Australia
Structural Grid Descriptors Predict Within-Task Solver Success on ARC-AGI
Jun 8, 2026Ayan PendharkarFunctional Code Solvers Solvers
Ekka: Automated Diagnosis of Silent Errors in LLM Inference
Jun 3, 2026Yile Gu, Zhen Zhang, Shaowei Zhu +4Diagnosis Large Language Models(Llms
University of Washington · 2Amazon Web Services.
Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States
May 8, 2026Yunho Choi, Jongwon Lim, Woojin Ahn +3Reinforcement Learning With Verifiable Reward Verifiable Rewards
Graduate School of Data Science Seoul National University · Computer Science and Engineering Seoul National University
Maximizing Rollout Informativeness under a Fixed Budget: A Submodular View of Tree Search for Tool-Use Agentic Reinforcement Learning
May 6, 2026Yuelin Hu, Zhenbo Yu, Zhengxue Cheng +2Group-Based Reinforcement Learning Tree Search
Shanghai Jiao Tong University · Shanghai Maritime University
RA-CMF: Region-Adaptive Conditional MeanFlow for CT Image Reconstruction
Apr 28, 2026Md Shifatul Ahsan Apurba, Md Selim, Jin ChenCone-Beam Computed Tomography Meanflow
Biomedical Informatics and Data Science University of Alabama at Birmingham Birmingham, AL, United States · Computer Science Florida Polytechnic University Lakeland, FL, United States
COVTrack++: Learning Open-Vocabulary Multi-Object Tracking from Continuous Videos via a Synergistic Paradigm
Mar 25, 2026Zekun Qian, Wei Feng, Ruize Han +1Multi-Object Tracking Open-Vocabulary Object Detection
School of Computer Science and Technology, Tianjin University, Tianjin, China · Department of Computer Science, City University of Hong Kong, Hong Kong SAR, China · Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology
FloydNet: A Learning Paradigm for Global Relational Reasoning
Jan 27, 2026Jingcheng Yu, Mingliang Zeng, Qiwei YeIntermediate States
Beijing Academy of Artificial Intelligence

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

Effective Does Not Mean Useful: Conditional Functional Substitutability for Redundancy and Scaling in Transformers

Jiaheng Chen, Jiaxing Li, Yucheng Xiao, Xinyong Cai+3

Harbin Institute of Technology, Shenzhen, China · Northeastern University, Shenyang, China · Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China+2

Redundancy Scaling Transformer Architectures Substitute Conceptual Reorganization Intermediate States

Open paper AskPDF arXiv

Modern neural networks scale predictably, yet the mechanisms behind these regularities remain unclear. Neural redundancy is typically characterized by component importance or representational similarity, both indirect proxies. We view redundancy as an input-conditioned, dynamic relation: intermediate computational states are functionally redundant when they induce similar downstream responses. We introduce Conditional Functional Substitutability (CFS) to directly characterize such functional substitution. CFS exposes functional relations and reduction potential missed by conventional importance- and similarity-based measures. Across modalities and Transformer families, CFS reveals systematic functional reorganization with scale. Controlled scaling further shows that performance gains need not track growth in substitutability, while fixed-capacity models with more independent functional structure perform better, providing a functional account of diminishing returns. Predicted CFS further enables dynamic computation with a better performance--computation trade-off than importance-based component selection, suggesting new directions for redundancy-aware computation and more efficient model scaling.

Momentum

Related topics

Latest papers 18