Reverse

Follow this topic Ask about this topic

Momentum

7 papers in the last four weeks, level with the four weeks before. 0.1% of all new papers.

Jul 6Week of Sep 21

Latest papers 58

All topics

RoPE at the End of Its Rope? Theory, Diagnosis, and Mitigation of Long-Context Failures
Sep 30, 2026Yuyang Wu, Yufeng Du, Hao PengEfficient Long-Context Inference Rotary Position Embedding
University of Illinois at Urbana-Champaign, USA
Is your uncertainty map wrong, or is its target? Exact diagnostics for the Tweedie diagonal, and a gradient-free alternative
Sep 27, 2026Vicent Ribas, Anna Oliveras TousDiagnosis Diffusion Models
Eurecat · University of Barcelona
Rethinking Token Reweighting for SFT: Suppress, Reverse, and Extrapolate Learned Features
Sep 27, 2026Cunchun Li, Haonan He, Yifan Gao +4Supervised Finetuning Reweighting
Shanghai AI Laboratory · University of Science and Technology of China · Fudan University +2
Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models
Sep 27, 2026Injin Kong, Sunghwan Choi, Yohan JoMasked Diffusion Language Models Diffusion Language Models
Graduate School of Data Science, Seoul National University
Benchmarking MLLMs via Cognitive Expected Scene Graph for Safety-Critical Visual Negation Understanding
Sep 17, 2026Zhiyun Jiang, Hanyong Wang, Binbin Liang +3Negation Situational Awareness
Sichuan University, Chengdu, China · Beijing Institute of Technology, Beijing, China
REVERSAL-BENCH: A Reversibility Axis and Reset Oracle for Measuring the Reset-Free RL Cliff
Sep 15, 2026Riyaaz Shaik, Chandru VenkataramanOffline Reinforcement Learning Reverse
Apple
Audio Deepfake Detection Using Temporal Coherence Analysis
Sep 8, 2026Justin D. Norman, Sarah BarringtonAudio Deepfake Detection Deepfake Detection
University of California, Berkeley
Corner Cases: Headland Coverage Path Planning for Autonomous Driving in Arable Farming
Sep 3, 2026Riikka Soitinaho, Timo OksanenPath Planning Precision Agriculture
Technical University of Munich, Germany; Chair of Agrimechatronics; Munich Institute of Robotics and Machine Intelligence (MIRMI)
Breaking the Compression Barrier: Cross-Architecture Compression Boundary Learning via Reverse Regrowth
Aug 17, 2026Zhaocen Liu, Satvik Praveen, Yi ShengCompressed Model Task-Aware Compression
University of British Columbia Vancouver, British Columbia, CA · University of South Florida Tampa, Florida, USA
Unifying Physical Backpropagation
Aug 12, 2026Cyrill Bösch, Yigithan Gediz, Clara C. Wanjura +1Backpropagation Adjoint Matching
Department of Computer Science, Princeton University, Princeton, NJ 08540, USA · Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ 08540, USA
On Weak Bisimilarities in CCSK
Aug 12, 2026Baptiste Vallée, Ivan LaneseReverse Similarity and Metrics
Ecole Normale Supérieure Paris-Saclay (France) · Olas Team, University of Bologna/INRIA (Italy)
TrustRoboReward: Preference-Ordered Isotonic Score Editing for Multi-Paradigm Robot Reward Models
Aug 9, 2026Yidong Wang, Yan Zhan, Ziteng Feng +16Progress Reward Modeling Reward Functions
Peking University · University of Science and Technology of China · Southeast University +6
Progressive Content Refinement with Decaying Reward Joint LinUCB
Aug 7, 2026Shion Ishikawa, Pablo Loyola, Young-joo Chung +1Contextual Bandit Framework Large Language Model Reinforcement Learning
Rakuten Group, Inc. Tokyo, Japan · Rakuten Group, Inc. San Mateo, USA
Velocity- and Regime-Aware Detection of Intraday Options Market Manipulation, with Explainable Attribution
Aug 5, 2026Alex Chen, Maria HybinetteVolatility Fraud Detection
University of Georgia · Athens, GA, USA
Caved or Convinced: Temporal Sampling Gates Claim Deference in Video Large Language Models
Aug 4, 2026Yuxin Cao, Wei Song, Jingling Xue +1Falsehood Reverse
National University of Singapore, Singapore · University of New South Wales, Australia
Evidence-Type Competition: When Can Interventional Data Teach Language Models Causal Direction?
Jul 31, 2026Xining XunCausal Intervention Reverse
Tsingjiao Information Science (Beijing) Co., Ltd.
Few-shot Deep Learning for Phase-Amplitude Aberration Correction in Transcranial Focused Ultrasound
Jul 31, 2026Minju Seol, Minjee Seo, Seonaeng Cho +1Upsampling Microphone Array
School of Mathematics and Computing (Computational Science and Engineering), Yonsei University, 03722, Seoul, Republic of Korea · Innovative & Intelligent Computational Science Institute (IN2CSI), Seoul, Republic of Korea
RAID: Towards Robust AI-Generated Image Detection with Bit-Reversed Images
Jul 31, 2026Renxi Cheng, Jie Gui, Hongsong WangAi-Generated Image Detection Data Generation
School of Cyber Science and Engineering, Southeast University, Nanjing 210096, China · Purple Mountain Laboratories, Nanjing 210000, China · Engineering Research Center of Blockchain Application, Supervision And Management (Southeast University), Ministry of Education, China +2
StanceFlip: A Comprehensive Multi-Dimensional Benchmark for Multimodal Conversational Stance Flipping Forecasting
Jul 27, 2026Heyan Chai, Xin Li, Wenjie Wang +5Multimodal Sentiment Analysis Flips
College of Computer Science and Software Engineering, Shenzhen University Shenzhen, China · Harbin Institute of Technology Shenzhen, China · Shenzhen University Shenzhen, China +1
Learning Forward & Reverse Skills from a Single Unfinished Demonstration for Constrained Manipulation Tasks
Jul 15, 2026Yexin Hu, Haoyi Zheng, Johannes Heidersberger +1Real-World Manipulation Tasks Motion Primitives
Autonomous Systems Lab, Technische Universität Wien (TU Wien), Austria · Institute of Robotics and Mechatronics, German Aerospace Center (DLR), Germany
Reverse to Advance: Teleoperation-Cost Effective Hard Policy Learning from Reversed Easy Tasks
Jul 15, 2026Qiyuan Qiao, Ge Yuan, Can Wang +1Teleoperation Robotic Manipulation Policies
The University of Hong Kong
What Does a Temporal Benchmark Score Measure? Decomposing Channel Use in Video VLM Evaluation
Jul 14, 2026Farrukh RahmanReverse Positional Encoding
Microsoft · Georgia Institute of Technology
Natural Ungrokking: Asymmetric Control of Which Rules Survive Pretraining
Jun 24, 2026Juliana Li, Diya SreedharReverse Pretraining
Harvard University, Cambridge, MA, USA.
EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis
Jun 24, 2026Huaqiu Li, Jiahao Wang, Sijia Cai +4Stylization Large-Scale Image Datasets
The Hong Kong University of Science and Technology · Alibaba Group · Xi’an Jiaotong University
When Does a Video-Language Model Stop Watching? Reward Strength Controls the Formation and Reversal of Visual Shortcuts in Multimodal RLVR
Jun 20, 2026Zekun XuReinforcement Learning With Verifiable Reward Video-Language Models
School of Computer Science, The University of Sydney, Sydney, Australia
NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment
Jun 16, 2026Jisung Hwang, Yunhong Min, Jaihoon Kim +2Diffusion Sampling Reward Gradients
1KAIST · †This work was done in part while the author was a visiting researcher at The University of Tokyo. · 2The University of Tokyo
SoK: AI-Augmented Binary Reversing
Jun 16, 2026Yujeong Kwon, Yiyue Zhang, Shakhzod Yuldoshkhujaev +3Reverse Engineering Bits
Sungkyunkwan University · The University of Chicago · Yonsei University
Incentives and Evidence in Learned Service Orchestration
Jun 15, 2026Syed Izhan Khilji, Alireza Furutanpey, Schahram DustdarOrchestration Kubernetes
†Distributed Systems Group (DSG), TU Wien, Vienna, Austria · ‡Coovally.ai, Barcelona, Spain
The Reverse Telescoping Coordinate System for Positive Definite Matrices: Geometry, Computation, and Generative Modeling
Jun 13, 2026Anindya BhadraMatrix Generative Models
Department of Statistics, Purdue University
RogueAI: A Reverse Turing Test for Detecting Licensed AI Deception in Dialogue
Jun 11, 2026Sara Candussio, Emanuele Ballarin, Lorenzo Bonin +2Deception Trustworthy Artificial Intelligence
1AILab, MIGe, University of Trieste, 34127 Trieste, IT · 2Computational Statistics and Machine Learning, Istituto Italiano di Tecnologia, Genova, IT · 3DIA, University of Trieste, 34127 Trieste, IT
Strategic Decision Support for AI Agents
Jun 10, 2026Shayan Kiyani, Sima Noorani, George Pappas +1Software Engineering Decision Support Artificial Intelligence Agents
University of Pennsylvania
RPO-PDT: Demonstrating Role-Play-Based Knowledge Adaptation for Student Support Dialogue (Demonstration System)
Jun 8, 2026Filip Janik, Ewa Olton, Robert Smales +4Tutors Role-Playing Agents
Edinburgh Napier University
Reversible Foundations: Training a 120B Sparse MoE through State-Preserving Scaling
Jun 5, 2026Rohan ShravanMixture-Of-Experts Architectures Large Language Model Training
The School of AI Bengaluru, India
Evidence Graph Consistency in Retrieval-Augmented Generation: A Model-Dependent Analysis of Hallucination Detection
Jun 4, 2026Jianru ShenCitation Hallucination Detection Hallucination Detection
University of Montana Missoula, MT, USA
ReverseEOL: Improving Training-free Text Embeddings via Text Reversal in Decoder-only LLMs
Jun 4, 2026Ailiang Lin, Zhuoyun Li, Yusong Wang +3Frozen Language Model Reverse
Institute of Science Tokyo · Tencent
Stability vs. Manipulability: Evaluating Robustness Under Post-Decision Interaction in LLM Judges
Jun 3, 2026Srimonti Dutta, Akshata Kishore MoharirLlm-As-A-Judge Large Language Model Judges
WAI USA Research Labs
Multilingual Unlearning in LLMs: Transfer, Dynamics, and Reversibility
Jun 2, 2026Chaoyi Xiang, Olga Ohrimenko, Benjamin I. P. Rubinstein +1Large Language Model Unlearning Reverse
School of Computing and Information Systems, The University of Melbourne, Melbourne, Australia.
Conditional Hypothesis Generation for LLM-Based Text Analysis with Researcher-Specified Covariates
Jun 2, 2026Paiheng Xu, Jing Liu, Wei AiComputational Social Science Covariates
University of Maryland, College Park
Bayesian Spectral Emotion Transition Discovery from Multi-Annotator Disagreement
Jun 1, 2026Keito Inoshita, Takato UenoEmotion Recognition Transitions
Faculty of Business and Commerce, Kansai University, Suita, Japan · Data Science and AI Innovation Promotion Center, Shiga University, Hikone, Japan · Graduate School of Data Science, Shiga University, Hikone, Japan
Reverse Probing: Supervised Token-level Uncertainty Quantification for Large Language Models in Clinical Text
May 27, 2026Bushi Xiao, Sarvesh Soni, Daisy Zhe WangLarge Language Model Uncertainty Token-Level Uncertainty
University of Florida Gainesville, FL · U.S. National Library of Medicine Bethesda, MD
How VLAs Fail Differently: Black-Box Action Monitoring Reveals Architecture-Specific Failure Signatures
May 27, 2026Krishnam GuptaVision-Language-Action Framework Diffusion-Based Vision-Language-Actions
Independent Research San Francisco, CA
ReverseMath: Answer Inversion for Scalable and Verifiable Mathematical Problem Generation
May 26, 2026Raoyuan Zhao, Yihong Liu, Yupei Du +2Mathematical Reasoning Benchmarks Reverse
Center for Information and Language Processing, LMU Munich · 2Munich Center for Machine Learning (MCML) · 3Saarland University
REVERSE: Reinforcing Evidence Verification and Search for Agentic Image geo-localization
May 26, 2026Yong Li, Furong Jia, Dacheng Yin +4Cross-View Geo-Localization Reverse
Peking University · The Hong Kong University of Science and Technology · WeChat Vision, Tencent Inc
SafetyRepro: Configuration-Conditional Rank Instability on Alignment Benchmarks
May 25, 2026Yanhang Li, Zhichao Fan, Zexin ZhuangReverse Disagreement
Northeastern University, Boston, MA, USA · University of Illinois Urbana-Champaign, Urbana, IL, USA · Southern Methodist University, Dallas, TX, USA
ReversedQ: Opportunities for Faster Q-Learning in Episodic Online Reinforcement Learning
May 20, 2026Sofia R. Miskala-Dinc, Aviva PrinsQ-Learning Model-Free Reinforcement Learning
University of Maryland College Park, MD, United States
Reversa: A Reverse Documentation Engineering Framework for Converting Legacy Software into Operational Specifications for AI Agents
May 18, 2026Sanderson Oliveira de Macedo, Ronaldo Martins da CostaReverse Engineering Documentation
Federal Institute of Goias - Brazil · Federal University of Goias - Brazil
Ablating Safety: Mechanisms for Removing Alignment in Language Models for Security Applications
May 17, 2026Isaac David, Arthur GervaisLarge Language Model Safety Security Evaluation
University College London
Support Before Frequency in Discrete Diffusion
May 13, 2026Adrian Müller, Antoine Gonon, Zebang Shen +2Frequency Iterative Denoising Methods
Department of Computer Science, ETH Zürich, Switzerland · Institute of Mathematics, EPFL, Switzerland · Institute of Statistical Science, Academia Sinica, Taiwan
The Growing Pains of Frontier Models: When Leaderboards Stop Separating and What to Measure Next
May 13, 2026Adil AminFrontier Models Leaderboard
ZEHEN Labs
Fair outputs, Biased Internals: Causal Potency and Asymmetry of Latent Bias in LLMs for High-Stakes Decisions
May 12, 2026Jagdish Tripathy, Marcus BuckmannLarge Language Model Bias High-Stakes Domains
Bank of England
Same Signal, Opposite Meaning: Direction-Informed Adaptive Learning for LLM Agents
May 7, 2026Ziming Li, Jiatan Huang, Xiaoguang Guo +2Large Language Model Agents Gating
University of Connecticut · New Jersey Institute of Technology
Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse Dynamics
May 7, 2026Muhammad H. Ashiq, Samanyu Arora, Abhinav N. Harish +3Diffusion Sampling Denoising Diffusion Probabilistic Model
University of Wisconsin-Madison, Madison, WI, USA.
Hybrid A*-Based Reverse Path-Planning of a Vehicle with Trailer System
Apr 27, 2026Xincheng Cao, Haochong Chen, Bilin Aksun-Guvenc +6Collision Avoidance Path Planning
Automated Driving Lab, Department of Mechanical and Aerospace Engineering, The Ohio State University, Columbus, OH 43210, USA · Hyundai America Technical Center, Inc. (HATCI), Superior Township, MI 48198, USA
Robust Continual Unlearning against Knowledge Erosion and Forgetting Reversal
Apr 21, 2026Eun-Ju Park, Youjin Shin, Simon S. WooMachine Unlearning Forgetting
1Sungkyunkwan University · 2The Catholic University of Korea · 3Secure Machines Lab
Reverse Constitutional AI: A Framework for Controllable Toxic Data Generation via Probability-Clamped RLAIF
Apr 20, 2026Yuan Fang, Yiming Luo, Aimin Zhou +1Adversarial Prompts Toxicity
East China Normal University, Shanghai, China · Shanghai Innovation Institute, Shanghai, China
Rewind-IL: Online Failure Detection and State Respawning for Imitation Learning
Apr 17, 2026Gehan Zheng, Sanjay Seenivasan, Matthew Johnson-Roberson +1Imitation Learning Reverse
College of Connected Computing, Vanderbilt University, USA. · School of Computer Science, University of Waterloo, Canada. · School of Computer Science, The University of Sydney, Australia. +1
Neural Continuous-Time Markov Chain: Discrete Diffusion via Decoupled Jump Timing and Direction
Apr 17, 2026Jingyuan Li, Xiaoyi Jiang, Fukang Wen +5Diffusion Models Language Model Perplexity
2Beijing Institute of Mathematical Sciences and Applications · 3Wuhan University · 4MathonAI +1
Fast Convergence for High-Order ODE Solvers in Diffusion Probabilistic Models
Date pendingDaniel Zhengyu Huang, Jiaoyang Huang, Zhengjiang LinTime Discretization Ordinary Differential Equations
Beijing International Center for Mathematical Research, Center for Machine Learning Research, Peking University, Beijing, China · Department of Statistics and Data Science, University of Pennsylvania, Philadelphia, PA, USA · Department of Mathematics, Massachusetts Institute of Technology, 77 Massachusetts Ave, 02139 Cambridge MA, USA

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

RoPE at the End of Its Rope? Theory, Diagnosis, and Mitigation of Long-Context Failures

Yuyang Wu, Yufeng Du, Hao Peng

University of Illinois at Urbana-Champaign, USA

Efficient Long-Context Inference Rotary Position Embedding Reverse Mitigation Theory

Open paper AskPDF arXiv

Long-context failures of RoPE-based language models can arise from RoPE's intrinsic tradeoff between maintaining stable token preferences and distinguishing nearby positions. Determining which weakness to address, and how, requires a more precise characterization of RoPE's behavior in trained models across context lengths. We address a key limitation of prior theory by allowing unequal query-key scales across RoPE frequencies, which aligns well with practical empirical observations. Our theory makes both vulnerabilities measurable for individual heads and inputs, and quantifies how high-frequency components support positional sensitivity while potentially disrupting semantic stability. We also derive a theoretical context-length bound beyond which, under specified conditions, a fixed attention-score comparison cannot jointly avoid semantic reversal and positional insensitivity. Guided by our fresh theoretical insights, we introduce RoPE Profiler, a lightweight, plug-and-play diagnostic toolkit that augments existing evaluations with zero additional forward passes by reusing cached query and key activations. Reusing activations collected during evaluation, the toolkit incurs little overhead. It supplements standard benchmark scores with two diagnostic scores that reveal semantic and positional weaknesses and help users prioritize which aspect to address. Crucially, our evaluations across 49 long-context task settings reveal a distinct pattern where reasoning tasks predominantly suffer from semantic reversal, whereas retrieval tasks are primarily vulnerable to positional insensitivity. Guided by our theory and diagnostic profiles, targeted high-frequency rescaling achieves immediate gains without additional training, improving task accuracy by up to 20 percentage points on Qwen3-8B and 25 percentage points on Llama-3.1-8B-Instruct.

Momentum

Related topics

Latest papers 58