Search

Topics

Multi-Object Tracking243 papers Eye Tracking45 papers Unmanned Aerial Vehicle Tracking45 papers 3D Single Object Tracking25 papers Catheter Tip Tracking11 papers Search Algorithms91 papers Motion Estimation35 papers

Papers 3,571–3,600 of 3,611

From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning
Jun 16, 2026Lingjing Kong, Xin Liu, Guangyi Chen +9Large Language Model Reasoning Reasoning Traces
Carnegie Mellon University · Mohamed bin Zayed University of Artificial Intelligence · Institute of Foundation Models +1
Otters++: A Time-to-first-spike Based Energy Efficient Optical Spiking Transformer
Jun 11, 2026Zhanglu Yan, Jiayi Mao, Kaiwen Tang +6Spiking Neural Networks Binary Spikes
Department of Computer Science, National University of Singapore · School of Engineering, Westlake University · College of Computer Science and Technology, Zhejiang University +2
Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation
Jun 4, 2026Maxime Griot, Paul Steven Scotti, Tanishq Mathew AbrahamKnowledge Distillation Reasoning Traces
Université catholique de Louvain · Sophont Inc.
Failed Reasoning Traces Tell You What Is Fixable (But Not by Reading Them)
Jun 3, 2026Nizar Islah, Istabrak Abbes, Irina Rish +2Reasoning Failures Reasoning Traces
1Mila - Quebec AI Institute · 2Université de Montréal · 3Polytechnique Montréal +1
Consensus is Strategically Insufficient: Reasoning-Trace Disagreement as a Knowledge-Representation Signal
Jun 2, 2026Michał Wawer, Jarosław A. ChudziakMulti-Agent Reasoning Disagreement
Laboratory of The New Ethos Warsaw University of Technology, Warsaw, Poland · Institute of Computer Science Faculty of Electronics and Information Technology Warsaw University of Technology, Warsaw, Poland
Reasoning over Grammar: Can Synthetic Linguistic Reasoning Traces Enhance Low-Resource Machine Translation?
Jun 2, 2026Renhao Pei, Yihong Liu, Sampo Pyysalo +2Machine Translation Linguistics
ELLIS Institute Finland · University of Turku · Center for Information and Language Processing, LMU Munich +1
EURO-5K: When Does Domain Pretraining Matter? Benchmarking Transformers for EU Reporting Obligation Extraction
Jun 2, 2026Marios Koniaris, Vasileios Kotronis, Eugenia Giannini +1Text Corpora Report
Division of Computer Science, School of Electrical and Computer Engineering, National Technical University of Athens, Iroon Polytechniou 9, Zographou Campus, 15780, Athens, Greece. · Department of Humanities Social Sciences and Law, School of Applied Mathematical and Physical Sciences, National Technical University of Athens, Iroon Polytechniou 9, Zographou Campus, 15780, Athens, Greece.
Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs
May 30, 2026Yu-An Lu, Ci-Yang Tsai, Yu-Lin Tsai +2Reasoning Traces Large Language Model Reasoning
National Yang Ming Chiao Tung University · UC Berkeley
Conformal Certification of Reasoning Trace Prefixes
May 28, 2026Matt Y. Cheung, Ashok Veeraraghavan, Hanjie Chen +1Reasoning Traces Certification
Department of Electrical & Computer Engineering, Rice University · Department of Computer Science, Rice University
Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces
May 28, 2026Chen He, Yuhao Wu, Lei Wang +2Large Language Model Post-Training Continuation
University of Electronic Science and Technology of China · Singapore University of Technology and Design · Singapore Management University
ReasonOps: Operator Segmentation for LLM Reasoning Traces
May 28, 2026Daniel Lee, Owen Queen, James ZouReasoning Traces Large Language Model Reasoning
Stanford University
Explaining Too Much? Understanding How Large Language Model Reasoning Traces Influence Performance and Metacognition
May 25, 2026Daniela Fernandes, Daniel Buschek, Lev Tankelevitch +2Reasoning Traces Large Language Model Reasoning
Aalto University, Espoo, Finland · University of Bayreuth, Bayreuth, Germany · Microsoft Research Cambridge, Cambridge, United Kingdom +1
Rethinking Transfer Learning for Industrial Inspection: DINOv3 vs. ImageNet Pretraining Across RGB and X-ray Tasks
May 22, 2026Mehdi Gharbage, Céline Teulière, Pierre Bouges +1Dinov3 Imagenet
1Michelin Tyres Manufacturer · 2Universit´e Clermont Auvergne, CNRS, Institut Pascal, Clermont-Ferrand, France
Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models
May 20, 2026Dong Chen, Fangyun Wei, Ziyu Wan +18Text-To-Image Image-Text Pairs
Microsoft Lens Team
Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning
May 20, 2026Lukas Twist, Helen Yannakoudakis, Jie M. ZhangReasoning Traces
King’s College London London, UK
Plug-and-Play Spiking Operators: Breaking the Nonlinearity Bottleneck in Spiking Transformers
May 19, 2026Xinzhe Yuan, Xiang Peng, Bin Gu +1Spiking Neural Networks Neuromorphic Hardware
IASM, Harbin Institute of Technology · IASM, Harbin Institute of Technology, China · School of Artificial Intelligence, Jilin University +1
Benchmarking transferability of SSL pretraining to same and different modality segmentation tasks
May 18, 2026Jue Jiang, Harini VeeraraghavanSwin Self-Supervised
Department of Medical Physics, Memorial Sloan Kettering Cancer Center
Attacking the First-Principle: A Black-Box, Query-Free Targeted Mimicry Attack on Binary Function Classifiers
May 18, 2026Gabriel Sauger, Jean-Yves Marion, Sazzadur Rahaman +3Malware Detection Evasion
UL, CNRS, LORIA, Inria · ∗Universit´e de Lorraine, CNRS, LORIA, F-54000 Nancy, France · CNRS, UL, LORIA, Inria +2
Auditing Reasoning-Trace Memorization Claims after Unlearning with Head-Conditioned Canaries
May 17, 2026Yanhang Li, Zhichao Fan, Zexin ZhuangMemorization Deepseek
Northeastern University, USA · University of Illinois Urbana-Champaign, USA · Southern Methodist University, USA
Characterizing Learning in Deep Neural Networks using Tractable Algorithmic Complexity Analysis
May 15, 2026Pedram Bakhtiarifard, Sophia N. Wilson, Mahmoud Afifi +2Source-Channel Coding
Department of Computer Science, University of Copenhagen, Denmark · The American University in Cairo, Egypt
Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces
May 14, 2026Sanjoy Chowdhury, Dinesh ManochaReasoning Traces Reasoning Benchmark
University of Maryland, College Park, USA
When Reasoning Traces Become Performative: Step-Level Evidence that Chain-of-Thought Is an Imperfect Oversight Channel
May 12, 2026Wenkai Li, Fan Yang, Ananya Hazarika +2Reasoning Traces Chain-of-Thought Reasoning
Carnegie Mellon University · Fujitsu Research of America Inc.
Unpacking the Eye of the Beholder: Social Location, Identity, and the Moving Target of Political Perspectives
May 11, 2026Elena SirotkinaPolitical Discourse Audience Engagement
Center for Data Science New York University
Benchmarking Transformer and xLSTM for Time-Series Forecasting of Heat Consumption
May 10, 2026Marja Wahl, Daniel R. Bayer, Sven Rausch +1Electricity Price Forecasting Bidirectional Long Short-Term Memory
RAUSCH Technology GmbH, Würzburg, Germany · Modeling and Simulation, University of Würzburg, Würzburg, Germany
Uncertainty-Aware Token Importance Estimation in Spiking Transformers
May 10, 2026Wenxuan Liu, Zecheng Hao, Tong Bu +2Transformer Architectures Importance
School of Computer Science, Peking University · School of Computer Science, Peking University. · Institute for Artificial Intelligence, Peking University +2
VecCISC: Improving Confidence-Informed Self-Consistency with Reasoning Trace Clustering and Candidate Answer Selection
May 8, 2026James Petullo, Sonny George, Dylan Cashman +1Reasoning Traces Self-Consistency
Computer Science Department, Brandeis University
SAFformer:Improving Spiking Transformer via Active Predictive Filtering
May 8, 2026Zequan Xie, Weiming Zeng, Yunhua Chen +3Spiking Neural Networks Hybrid Transformer
School of Computer Science and Technology, Guangdong University of Technology, Guangzhou, China · Faculty of Science, Hong Kong Baptist University, China · School of Electronic Information, Wuhan University, China
Extracting Search Trees from LLM Reasoning Traces Reveals Myopic Planning
May 7, 2026Sixing Chen, Ji-An Li, Saner Cakir +3Large Language Model Planning Large Language Model Reasoning
Department of Psychology New York University · Generality Inc.
Distributional Spectral Diagnostics for Localizing Grokking Transitions
May 7, 2026Ziyue Wang, Yufeng Ying, Takafumi KanamoriGrokking Modular Addition
Institute of Science Tokyo · University of Science and Technology of China
Elastic Spiking Transformers for Efficient Gesture Understanding
May 4, 2026Alberto Ancilotto, Gianluca Amprimo, Stefano Di Carlo +1Spiking Neural Networks Neuromorphic Hardware
Fondazione Bruno Kessler · Politecnico di Torino

Jun 16, 2026, cs.LGJ/K move · Enter open · S save

From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning

Lingjing Kong, Xin Liu, Guangyi Chen, Martin Q. Ma+8

Carnegie Mellon University · Mohamed bin Zayed University of Artificial Intelligence · Institute of Foundation Models+1

Large Language Model Reasoning Reasoning Traces Recursive Reasoner Reuse

Open paper AskPDF arXiv

Post-training pipelines that combine supervised fine-tuning (SFT) with reinforcement learning (RL) have emerged as the key recipe for transforming large language models (LLMs) into robust reasoners. We argue that this combined success is driven by compositional generalization, which we formalize through a hierarchical latent selection model. In this framework, reasoning traces are generated by a cascade of discrete latent selection variables corresponding to reusable atomic modules, including both skills (local operations) and routing mechanisms (how intermediate information is selected, reused, and composed). Within this model, we theoretically show that SFT and RL play asymmetric, complementary roles: SFT supplies the raw module materials in compositional traces, and RL decomposes those traces to identify the latent atomic modules and enable compositional generalization. We design controlled experiments to validate this theory. Our results demonstrate that RL can extract atomic modules from compound traces supplied by SFT and recombine them to solve new configurations. Moreover, we find that training on compound traces yields stronger generalization than training on isolated atomic modules. Finally, we investigate the relationship between SFT and RL data and identify an effective protocol in which SFT ensures coverage of all atomic modules through compositional traces, while RL focuses on novel compositions outside the SFT support to drive exploration.