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
[abs]Speak Early: Speculative Speech Generation for Low-Latency SpeechLLMsTianyuan Jiang, Yongheng Deng, Ju Ren4.00
[abs]WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory LearningSang-Hoon Lee, Ha-Yeong Choi4.67
[PDF]Rethinking Vector Field Learning for Generative SegmentationChaoyang Wang, Yaobo Liang, Boci Peng et al.4.33
[abs]AEGIS: Almost Surely Safe Offline Reinforcement LearningJunseo Lee, Hyeokjin Kwon, Songhwai Oh4.33
[PDF]EO-WM: A Physically Informed World Model for Probabilistic Earth Observation ForecastingJunwei Luo, Shuai Yuan, Zhenya YANG et al.4.00
[PDF]LongScape: Advancing Long-Horizon Embodied World Models with Context-Aware MoELei Jin, Yu Shang, Yiding Ma et al.3.67
[abs]Two Phase Rapid Simulation Based Inference with Differentiable SimulatorsUros Zivanovic, Rahul Srinivasan, Roberto Trotta et al.5.00
[abs]Metric Depth Estimation from Arbitrarily Degraded Low-Resolution Depth PromptsKun Wang, Yun Zhu, Pan Zhou et al.4.33
[PDF]Discrete Stochastic Localization for Non-autoregressive GenerationYunshu Wu, Jiayi Cheng, Longxuan YU et al.4.20
[PDF]Baton: Explicit Semantic Blueprints for Joint Video-Audio GenerationShuyuan Tu, Qi Tian, Zihan Yang et al.4.75
[abs]High-Fidelity Boltzmann Samplingvia Physical Prior Lifted Continuous GFlowNetsXizhi Tian, Wenhao Deng, Haojia Hui et al.4.00
[abs]FlashControl: One-Step Controllable Generator via Distillation-Friendly Single-Stream Teachers.Ngan Nguyen, Dung Nguyen, Quan Dao et al.4.67
[PDF]VideoMDM: Towards 3D Human Motion Generation From 2D SupervisionAmir Mann, Gal M Harari, Merav keidar et al.4.33
[abs]Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-TuningJintao Zhang, Kai Jiang, Chendong Xiang et al.4.67
[PDF]Rethinking Causal Action Tokenization with Conditional Annealing in Flow MatchingChenyu Zhang, Yuhang Cao, Yingxi Lu et al.4.33
[PDF]Embedding Security Properties into AI-Enabled Cyber-Physical SystemsZiyan An, John Stankovic, Meiyi Ma3.50
[PDF]PhysFormer: Learning to Simulate Mechanics in World SpaceYiming Chen, Yushi LAN, Andrea Vedaldi4.25
[PDF]Optimal Convergence Analysis of DDPM for General DistributionsYuchen Jiao, Yuchen Zhou, Gen Li5.00
[PDF]End-to-End Training for Unified Tokenization and Latent DenoisingShivam Duggal, Xingjian Bai, Zongze Wu et al.4.67
[abs]Bernoulli Flow Models: Self-Consistent Generative Modeling for Binary DataHao Mo, Liying Yang, Shumin Yao et al.3.67
[PDF]Learning Visual Feature-Based World Models via Residual Latent ActionXinyu Zhang, Zhengtong Xu, Yutian Tao et al.4.33
[PDF]From Representational Complementarity to Dual Systems: Synergizing VLM and Vision-Only Backbones for End-to-End DrivingSining ANG, Yuguang Yang, Chenxu Dang et al.4.00
[PDF]Calibrating Generative Models to Feature Distributions with MMD FinetuningNathaniel L. Diamant, Brian L Trippe4.33
[abs]MaRK: Markov-adapted Recurrent Kernels for Dynamic Operator Conditioning in State Space ModelsSyed Ibrahim Omer, Ginny Wong, Xiangyu Zhao4.00
[PDF]X2HDR: HDR Image Generation in a Perceptually Uniform SpaceRonghuan Wu, Wanchao Su, Kede Ma et al.4.67
[abs]FreqCa: Accelerating Image generation and editing via Frequency-Aware CachingJiacheng Liu, Peiliang Cai, Qinming Zhou et al.4.00
[PDF]Equilibrium Matching: Generative Modeling with Implicit Energy-Based ModelsRunqian (Ray) Wang, Yilun Du4.00
[abs]Distilling Conditional Image Generators into Spatial Effect MapsEun Ryung Lee, Seyoung Park4.00
[PDF]SteerVTE: Seamless Video Text Editing with Style and Glyph ControlKai Zeng, Moran Li, Zhengwei Wang et al.4.00
[abs]Learnable Chernoff Baselines for Provable Inference-Time AlignmentSunil Madhow, Yuchen Liang, Ness Shroff et al.3.33
[PDF]DriveFuture: Future-Aware Latent World Models for Autonomous DrivingYufeng hong, Xiaotian Zhou, Yingyan Li et al.4.33
[PDF]Effective Multi-sensor Conditioning for Street-view Novel-view SynthesisZhengfei Kuang, Adam Sun, Liyuan Zhu et al.4.67
[PDF]ExtraVAR: Stage-Aware RoPE Remapping for Resolution Extrapolation in Visual Autoregressive ModelsFeihong Yan, Shaoyu Liu, Haixuan Wang et al.4.33
[abs]LiteNav: Lightweight Map-free Outdoor Visual NavigationMingxu Zhu, Chang Liu, Xiao Zhao et al.4.00
[PDF]FASTER: Value-Guided Sampling for Fast RLPerry Dong, Alexander Swerdlow, Dorsa Sadigh et al.3.67
[PDF]STREAM: Stochastic Riemannian Flow Matching with Anisotropic Decoder for Digital Histopathology Image GenerationWon June Cho, Daeky Jeong, Hyeongyeol Lim et al.4.25
[abs]Agentic Video Editing from Underspecified RequestsYongsheng Yu, Ziyun Zeng, Zhiyuan Xiao et al.4.33
[PDF]InstructVVT: Instruction-Driven Video Virtual Try-On without Auxiliary Spatial PriorsDingbao Shao, Song Wu, Xinyu Chen et al.3.67
[PDF]Improved Baselines with Representation AutoencodersJaskirat Singh, Boyang Zheng, Zongze Wu et al.4.33
[PDF]Modeling Whole-Slide Images as Dynamic Tumor Microenvironment FieldsLei Wu, Jiashuai Liu, Di Zhang et al.4.00
[abs]Generative Active Learning via Bayesian Acquisition for Improving the Efficiency of Synthetic DataJeongjun Lee, Gyeonghoon Ko, Juho Lee4.33
[PDF]PACE: Phase-Aware Chunk Execution for Robot Policies with Action ChunkingJunnan Nie, Jiayi Li, Jiachen Zhang et al.4.00
[abs]Seeing the Unseen: Unified Visible–Invisible Motion for Physically Consistent Video GenerationMingyang Bao, Fangda Ye, Xihao Chen et al.4.00