NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of NeurIPS 2026
44 results for “"reward hacking" OR "specification gaming"” in NeurIPS 2026. clearsort:relevanceratingnewestmost read
[PDF]Are we really tilting? The mechanics of reward guidance in flow and diffusion modelsSanjit Dandapanthula, Nicholas Boffi3.33
[abs]Model Incrimination: Investigating Whether Concerning Behavior Reflects MisalignmentGerson Kroiz, Aditya Singh, Senthooran Rajamanoharan et al.5.00
[abs]RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with LLMsXiao Feng, Bo Han, Zhanke Zhou et al.4.00
[PDF]IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable RewardsXu Guo, Tianyi Liang, Jian Tong et al.4.33
[PDF]Verifier-Backed Hard Problem Generation for Mathematical ReasoningYuhang Lai, Jiazhan Feng, Yee Whye Teh et al.4.00
[abs]From Representation to Intervention: Using Emotion Vectors to Monitor and Guide Language ModelsGeorgia Dimaki, Martin Villanueva Dimakis4.00
[abs]Locality-Controlled OOD Guidance for Robust Regulatory DNA Sequence DesignDabin Jeong, Marwane Bourdim, Mohammad Lotfollahi4.50
[abs]Tail Competition Explains Scaling in Best-of-N Sampling for Verifiable ProblemsTaro Yano, Yoichi Ishibashi, Masafumi Oyamada4.00
[PDF]SR-GRPO: Stable Rank as an Intrinsic Geometric Reward for Large Language Model AlignmentYixuan Tang, Yi Yang4.00
[PDF]Visual-ERM: Reward Modeling for Visual EquivalenceZiyu Liu, Shengyuan Ding, Xinyu Fang et al.4.33
[PDF]OdysSim: Building Foundation Models for Human Behavior SimulationXuhui Zhou, Weiwei Sun, Weihua Du et al.4.33
[PDF]GARDO: Reinforcing Diffusion Models without Reward HackingHaoran He, Yuxiao YE, Jie Liu et al.4.00
[abs]Auto-Rubric as Reward: From Implicit Preference to Explicit Generative CriteriaJuanxi Tian, Fengyuan Liu, Jiaming Han et al.4.00
[abs]MemeEconomy : Do LLM Agents Trade Ethics for Survival?Syed Nazmus Sakib, Nafiul Haque, Ahnaf Manan et al.4.33
[abs]Pareto Preference Optimization for Structure- and Stability-Aware RNA Inverse FoldingMinghao Sun, Hanqun Cao, Zhou Zhang et al.4.00
[abs]Efficient Off-Policy RL for Video Generation via Forward-Consistent Reward MatchingHongzheng Yang, Mengyang LIU, Haoxuan Wu et al.4.67
[abs]GPA: Generative Population Annealing for Test-Time Sequence Design with Pretrained Generative ModelsAnirban Sarkar, Alejandra Duran, Peter Koo4.00
[abs]BootstrapAgent: Turning Repository Setup into Reusable Agent KnowledgeSihan Fu, Oucheng Liu, Shiyuan Wang et al.4.33
[PDF]SWE-Marathon: Can AI Agents Autonomously Complete Ultra-Long-Horizon Software Work?Rishi Desai, Joan Cabezas, Neel Harsola et al.
[PDF]Reinforcement World Model Learning for LLM-based AgentsXiao Yu, Baolin Peng, Ruize Xu et al.4.00
[PDF]Training on Documents About Monitoring Leads to CoT ObfuscationReilly Haskins, Bilal Chughtai, Joshua Engels4.33
[PDF]OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search FrameworkBen Chen, Siyuan Wang, Yufei Ma et al.3.67
[PDF]G-Zero: Self-Play for Open-Ended Generation from Zero DataChengsong Huang, Haolin Liu, Tong Zheng et al.4.00
[PDF]TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion AlignmentJiaming Li, Chenyu Zhu, Zhiyuan Ma et al.4.33
[PDF]SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware LimitsEdward C Lin, Sahil Modi, Siva Kumar Sastry Hari et al.
[abs]Designing Effective Monitor-Based Interventions for Mitigating Reward Hacking During RLAria Wong, Joshua Engels, Neel Nanda4.67
[abs]Gradient Routing Localizes and Removes Unintended Behaviors in RLJake Ward, Shawn Hu, Aria Wong et al.4.67
[PDF]-GRPO & Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM AlignmentRajdeep Haldar, Lantao Mei, Guang Lin et al.4.00
[abs]The VLM as Sensor: Bayesian Active Search for Long Video UnderstandingChong Tang, Sannara EK, Dirk Koch et al.4.33
[PDF]Theoretical Limits of Language Model AlignmentLucas Monteiro Paes, Natalie Mackraz, Barry-John Theobald et al.4.67
[PDF]Test-Time Personalization: A Diagnostic Framework and Probabilistic Fix for Scaling FailuresLinhai Zhang, Yulan He4.33
[PDF]SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement LearningPhilip Schroeder, Thomas Weng, Karl Schmeckpeper et al.4.67
[PDF]Reward Hacking in Rubric-Based Reinforcement LearningAnas Mahmoud, MohammadHossein Rezaei, Zihao Wang et al.4.67
[PDF]Two is better than one: A Collapse-free Multi-Reward RLIF Training FrameworkShourov Joarder, Diganta Sikdar, Ahsan H Akash et al.3.67
[PDF]Imperfect World Models are ExploitableLogan M Bhamidipaty, Esmeralda S Whitammer, David Abel et al.4.67
[PDF]Manifold Drift in Flow Preference Optimization: A Root Cause of Reward HackingYansen Han, Shengyi Liao, Yuanxing Zhang et al.4.00
[abs]Heuresis: Evaluating Search Strategies for Autonomous Machine Learning Research AgentsAntonis Antoniades, Deepak Nathani, Ritam Saha et al.4.00
[abs]StainNFT: Curriculum-Gated Multi-Reward Post-Training for Pathology-Faithful Virtual Stainingshurong yang, Dong Wei, Qiong Peng et al.5.00
[abs]Alignment Needs 'Cognitive Control': On The Role of Regularization in LLM AlignmentNona Ghazizadeh, Elnaz Rahmati, Morteza Dehghani et al.
[PDF]Hack-Verifiable Environments: Towards Evaluating Reward Hacking at ScaleAmit Roth, Ankur Samanta, Matan Halevy et al.