diffusion model
Imagine you take a clear image and slowly add random noise until it becomes pure static. A diffusion model learns the reverse: starting from random static, it gradually removes noise, step by step, to generate a new, clean image. This same idea works for text, video, or even physical trajectories, the model learns to "denoise" corrupted data, producing realistic outputs from noise.
Across the papers, researchers tackle key weaknesses of diffusion models: memorization, inefficiency, and poor generalization. They reinterpret denoising as attractor dynamics to detect overfitting, fix retry loops in embodied agents by reversing the scoring direction, and learn optimal noise schedules through game theory. Others split generation into phases for faster distillation, correct deterministic drift in video, and scale routing solvers to huge instances with reinforcement learning. These innovations improve reliability, speed, and adaptability without extra inference cost.
- Papers
- 199
- Released code
- 3
- First seen
- Sept 2026
- Latest
- Nov 2026
199 papers in the last two months, against 0 in the two before.
Start here
Three papers to read first, in the order they came out- 1Bottlenecked diffusion models predict best when targeting clean data, not noise
Prediction target and loss space choices in diffusion models affect generation quality due to architectural bottlenecks imposing irreducible error floors.
Nov 2026 - 2Sped-up AI image generators lose diversity at a single layer
Distilled generators collapse diversity because they concentrate activation variance into fewer directions at a single intermediate block.
Nov 2026 - 3Image generators’ copying is linked to a few attention heads
Memorization in diffusion models is driven by a few shallow-layer attention heads with abnormally strong semantic control.
Nov 2026
How fast it is moving
Often studied with
flow matchingA generative training method that learns an ordinary differential equation mapping noise to data by matching velocity fields along probability paths.51 togetherworld model for agentsA predictive environment model that agents use to simulate outcomes, enabling planning, reasoning, and long-horizon task execution.23 togetherreinforcement learningA machine learning method where an agent learns to make sequences of decisions by receiving rewards or penalties for its actions.21 togetherWho is working on it
The papers
Most central to this idea first, not most recent.
- Top Universitycs.CV
Sped-up AI image generators lose diversity at a single layer
Distilled generators collapse diversity because they concentrate activation variance into fewer directions at a single intermediate block.
University of Illinois at Urbana-Champaign, Johns Hopkins University · Nov 2026
- Industrycs.LG
Bottlenecked diffusion models predict best when targeting clean data, not noise
Prediction target and loss space choices in diffusion models affect generation quality due to architectural bottlenecks imposing irreducible error floors.
POSTECH · Nov 2026
- Academiccs.CV
Image generators’ copying is linked to a few attention heads
Memorization in diffusion models is driven by a few shallow-layer attention heads with abnormally strong semantic control.
Southeast University, Engineering Research Center of Blockchain ApplicationSupervision and Management (Southeast University), Ministry of Education · Nov 2026
- Top Universitycs.LG
Random AI image generators now nearly as fast as non-random ones
Forward-value discretization speeds stochastic diffusion sampling by evaluating the model at a lookahead point instead of relying on higher-order approximations.
Chinese University of Hong Kong, University of Michigan · Nov 2026
- Big Techcs.CV
Video translation models maintain long-term consistency by matching input regions across frames
Matching regions in input video should produce matching outputs, enabling long-range consistency in real-time video-to-video translation.
Carnegie Mellon University, Adobe Research · Nov 2026
- Top Universitycs.CV
Lifting latent video models to pixels enables higher resolution and per-frame control
PixVid converts a pretrained latent video diffusion model to generate frames directly in pixel space, improving resolution, control, and latency.
Roblox, University of California, Merced · Nov 2026
- Top Universitycs.CV
Robots can generate continuous movements without a separate action network
Concertina unifies manipulation and cognition in a single diffusion vision-language model via continuous latent action tokens.
University of Sydney, TokenRhythm Technologies · Nov 2026
- Top Universitycs.LG
Diffusion model trajectories follow a universal geometric shape
Deterministic diffusion sampling trajectories share a universal hyperbolic geometric structure across data types and models.
UC Berkeley, Zhejiang University · Nov 2026
- Chinese Techcs.AI
Language models can generate text five times faster by drafting ahead
D3D uses a decoupled diffusion drafter for asynchronous speculative decoding, achieving average speedups of 5.00x and 5.71x.
Shanghai Innovation Institute, Shanghai Academy of AI for Science · Nov 2026
- Chinese Techcs.CV
Adaptive noise timing boosts video compression realism with 40% bitrate savings
Choosing the right moment to start denoising adaptively balances fidelity and realism in generative video compression.
Beihang University, Alibaba Group · Nov 2026
- Big Techcs.CV
Learning visual concepts before rendering improves image generation in controlled tests
Visual generation is better when autoregressive pretraining learns semantic tokens and diffusion post-training renders them.
University of California, San Diego, Adobe Research · Oct 2026
- Big Techcs.CV
Video generators can follow motion without distorting their starting noise
HME encodes motion into noise via Householder reflections, exactly preserving spatial Gaussianity and achieving a 2.7x speedup.
The Australian National University, Vertex Lab · Nov 2026
- Chinese Techcs.CV
Video generators improve longer clips by revisiting earlier generation steps
Using the intermediate denoising trajectory of prior chunks as context improves autoregressive long video generation quality.
Huawei Noah Research Institute · Nov 2026
- Top Universitycs.CV
Streaming video generators can now output frames at near-watchable speed
Streaming video generation can emit one finished chunk per concurrent forward round by pipelining denoising stages across model replicas, preserving quality while speeding throughput.
Shanghai Jiao Tong University, Xi’an Jiaotong University · Nov 2026
- Chinese Techcs.CV
Image editors can drift less while retaining most existing capabilities
TailSteer reduces cumulative drift in multi-turn image editing by applying preservation corrections exclusively in the final denoising steps.
MMLab, CUHK, SenseTime Research · Nov 2026
- Big Techcs.LG
AI models factor larger numbers, but costs grow exponentially
A transformer using discrete diffusion and beam search factors 68-bit semiprimes but scales exponentially with bit length.
Google DeepMind · Nov 2026
- Independentcs.LG
AI diffusion models can follow deterministic paths through discrete data
Exponential residual flows provide an ODE-based sampler for discrete diffusion models that preserves marginals and improves sampling.
Oct 2026
- Chinese Techcs.CV
Jointly generating object shapes and camera poses improves sparse-view reconstruction
A native 3D diffusion model jointly recovers object shape and camera motion from a few uncalibrated views.
The University of Hong Kong, Transengram · Oct 2026
- Big Techcs.CE
AI protein design improves by separating suggested changes from evaluation
Protein binder design improves when approximate gradients from a diffusion structure predictor propose sequence edits and full no-gradient predictor passes verify them, exploring multi-site then single-site.
Stanford University, HKUST (Guangzhou) · Nov 2026
- Big Techcs.CV
Image restoration models can share skills without full joint retraining
OptIMerge merges image-to-image diffusion models without retraining by learning structured weight combinations from few examples.
Google · Nov 2026
- Top Universitycs.CV
High-resolution image editing gets a 4x speedup by keeping a low-resolution anchor
FLASH enables faithful 2K/4K image editing by anchoring a high-resolution detail branch to a frozen 1K editing backbone.
Shanghai Innovation Institute, Fudan University · Nov 2026
- Academiccs.RO
Robots can borrow a video model's foresight and act 21 times faster
Compressing a video world model's internal features and distilling them into an action model lets a robot act about 21 times faster at run time without the video model.
Harbin Institute of Technology, Shenzhen · Nov 2026
- Top Universitycs.LGcode
Diffusion models can be composed and steered at inference time without retraining
Generalized Feynman-Kac Correctors sample from arbitrary combinations of pretrained diffusion models without retraining.
Tel Aviv University · Oct 2026
- Chinese Techcs.CV
Image generators can train over 140 times faster with pixel feedback
SpeeDiff enables end-to-end training of latent diffusion models using a Tweedie loss that prevents latent collapse, achieving dramatically faster convergence.
ByteDance Seed, California Institute of Technology · Oct 2026
- Top Universitycs.LG
Empirical denoiser enables privacy accounting for diffusion sampling
Privacy loss in diffusion sampling concentrates in the critical window where semantics form and can be certified and reduced by subsampling, early stopping, and hybrid public denoisers.
University of British Columbia, Stanford University · Oct 2026
- Top Universitycs.LG
Schedule-Free Optimization Beats EMA-Only Training in Large Diffusion Models
Large diffusion models require optimization that balances variance reduction with continued progress, beyond just exponential moving average.
Tsinghua University, Shanghai Qi Zhi Institute · Oct 2026
- Chinese Techcs.CV
BEV maps enable human-friendly control for embodied navigation video generation
Map2Vid reformulates navigation video generation as a BEV-map-to-video synthesis problem for intuitive control.
East China Normal University, King Abdullah University of Science and Technology (KAUST) · Oct 2026
- Independentcs.LG
AI weather forecasts can capture extreme rain without blurring storm patterns
FitzRoy uses reinforced diffusion to achieve state-of-the-art precipitation forecast skill and physical realism simultaneously.
Oct 2026
- Academiccs.CV
A controllable video model can serve as a photorealistic 3D avatar
Video Diffusion Avatars generate photorealistic multi-view consistent avatars from a single image using a controlled video diffusion model.
Technical University of Munich, Woven by Toyota · Oct 2026
- Research Labcs.CV
Image generators forget unwanted content more reliably while preserving image quality
ConceptACP uses sparse autoencoders to precisely remove target concepts from diffusion models without inference-time overhead.
Institute of Information Engineering, Chinese Academy of Sciences, Boston University · Oct 2026
- Big Techcs.CV
Small frozen-model corrector curbs drift in long autoregressive video generation
A 5.9M-parameter LoRA corrector eliminates drift in autoregressive video diffusion without real videos and no inference cost.
Seoul National University, Northeastern University · Oct 2026
- Big Techcs.CV
AI image editors can separately control how much each person changes
LOCI enables simultaneous multi-subject image editing with per-region continuous strength control and reduced cross-region leakage.
Meta AI · Oct 2026
- Chinese Techcs.CL
Progressive verification speeds up diffusion language model inference
PV-Decoding reduces sequential model forward passes by 8-37% while keeping the largest observed accuracy drop to 1.3% for diffusion language models.
Huawei Technologies, City University of Hong Kong · Oct 2026
- Research Labcs.CV
Two classifier heads align few-step diffusion with external rewards.
DualHeads adapts few-step diffusion generators using only external reward evaluations and reference samples via two lightweight classifier heads.
École Polytechnique, CNRS · Oct 2026
- Big Techcs.CV
3D diffusion model generates body CT masks with controllable anatomy, tumors, demographics
A 3D latent diffusion model generates whole-body CT masks with joint control over organ size, tumors, and demographics.
NVIDIA, National Institutes of Health · Oct 2026
- Top Universitycs.LG
Direct distribution rewards preserve diversity during diffusion model fine-tuning
Reward fine-tuning of diffusion and flow models directly regularizes the generated distribution via distribution-dependent per-sample rewards, improving diversity and reducing drift at matched reward.
University of Oxford, AITHYRA · Oct 2026
- Big Techcs.AI
Reinforcement learning adapts diffusion models to large routing problems
ResizeCO adapts a diffusion model trained on small routing instances to large TSP and CVRP instances via reinforcement learning, setting a new state of the art for short run times.
Bar-Ilan University, NVIDIA · Oct 2026
- Big Techcs.CL
Diffusion separates response timing from content in full-duplex speech
Diffuplex decouples when to speak from what to speak, enabling fast and intelligent full-duplex spoken dialogue.
NVIDIA · Oct 2026
- Top Universitycs.AI
Recursive latent reasoner boosts diffusion models for visual puzzles
A recursive reasoner coupled to a frozen diffusion model solves visual reasoning puzzles from pixels without symbolic supervision.
Warsaw University of Technology, IDEAS Research Institute · Oct 2026
- Academiccs.RO
Five human videos can adapt robot policies within five minutes
A 400M diffusion policy pretrained on 100 hours of egocentric human video learns new robot tasks from 5 demonstrations in under 5 minutes.
New York University · Oct 2026
- Big Techstat.ML
Shifting denoiser inputs enables guided sampling from discrete diffusion models
Internal guidance samples tilted discrete distributions by shifting and rescaling a single denoiser’s inputs instead of modifying its outputs.
Simons Institute, University of California Berkeley · Oct 2026
- Top Universitycs.RO
DexCast learns real-world dexterous control from human videos without teleoperation
DexCast learns dexterous robotic hand control from monocular human videos without teleoperation, achieving 87.5% success on real-world tasks.
Tsinghua University, TsingTec · Oct 2026
- Chinese Techcs.CV
Diffusion transformer depth profiles reflect prediction targets, not just feature granularity
The inverted-U depth profile in diffusion transformers reflects where the prediction target is encoded, not a coarse-to-fine semantic signal.
Sun Yat-sen University, Alibaba Group · Oct 2026
- Independentcs.CV
Parallel diffusion decoding rivals autoregressive visual grounding at 4B scale
GroundAnything uses blockwise diffusion to deliver faster parallel visual grounding while retaining strong localization accuracy across diverse tasks.
Oct 2026
- Chinese Techcs.CV
StreamErase enables real-time streaming video object removal at 26 FPS
StreamErase enables sub-second, real-time streaming video object removal at over 26 frames per second on a single GPU.
Wuhan University, Alibaba Group · Oct 2026
- Top Universitycs.LG
Memorization in diffusion models measured by persistence across noise scales
Diffusion models' fixed-scale denoiser dynamics reveal memorization through a critical noise scale.
Institute of Science and Technology Austria (ISTA) · Oct 2026
- Top Universitycs.CV
Predicting velocity fields helps video models generate more realistic fluids
FluidGen generates physically plausible fluid videos by jointly predicting a 3D velocity field alongside RGB.
Stanford University · Oct 2026
- Big Techcs.RO
One Model Predicts Robot Futures and Chooses Actions
SUAVE uses a masked diffusion transformer with discrete tokens to unify video, actions, and language into one model.
Columbia University, Toyota Research Institute · Oct 2026
- Academiccs.RO
Separating exploration and exploitation policies with classifier-free guidance improves diffusion RL fine-tuning
CFG-Policy combines separate exploration and exploitation diffusion policies via classifier-free guidance to improve RL performance while preserving behavioral diversity.
ShanghaiTech University, InstAdapt · Oct 2026
- Top Universitycs.CV
Two denoising experts make video diffusion distillation far more efficient
Glance Forcing distills bidirectional video diffusion into a few-step autoregressive model using one PF-ODE trajectory, outperforming a fully trained baseline at far lower collection cost.
Wuhan University, University of Electronic Science and Technology of China · Oct 2026
- Big Techcs.CV
Generative chroma keying from unaligned background references preserves foreground performance
FreeCHROMA generates composite videos from unaligned background references while preserving foreground identity and motion.
University of Maryland, College Park, Amazon · Oct 2026
- Chinese Techcs.LG
Reverse scoring helps diffusion agents escape repeated failed actions
Diffusion language model agents retry failed actions due to task-blind bias; reverse scoring corrects this without retraining.
Huawei Noah’s Ark Lab, Technical University of Darmstadt · Oct 2026
- Independentcs.CV
Adversarial post-training fixes high-frequency deficit in pixel diffusion models
Adversarial post-training improves pixel diffusion outputs by restoring missing high-frequency image statistics.
Oct 2026
- Chinese Techcs.CV
Adaptive reward routing improves joint audio-video diffusion training
Adaptive Reward Routing dynamically adapts update locations and reward weights, improving joint audio-video diffusion across multiple metrics.
MMLab@HKUST, The Hong Kong University of Science and Technology · Oct 2026
- Big Techcs.CV
Generative and agentic pipeline reconstructs scene-aligned 3D objects from casual photos
GATOR reconstructs complete, scene-aligned 3D objects with pose from one or more casual images using generative and agentic methods.
NVIDIA, Simon Fraser University · Oct 2026
- Top Universitycs.LG
Natural language can steer all-atom protein design across diverse objectives
Prompteina generates all-atom protein structures from natural language descriptions, outperforming baselines and experimentally validating binders.
University of Washington, National University of Singapore · Oct 2026
- Big Techcs.CV
Consistency error enables direct unlearning in few-step distilled diffusion models
Using consistency error instead of noise prediction error achieves effective unlearning in few step distilled text to image models.
Amazon AGI, Purdue University · Oct 2026
- Big Techcs.CV
Calibrated depth reward improves 4D consistency without harming visual quality
A single calibrated geometric reward improves both 4D consistency and video quality without auxiliary rewards.
KAIST, ETH Zürich · Oct 2026
- Academiccs.CV
Splitting diffusion distillation into two phase-specific half-sized experts beats a full-size student.
Phase-wise Velocity Distillation uses two half-sized phase-specific experts to generate high-quality images in one full-backbone forward pass.
The Hong Kong Polytechnic University, OPPO Research Institute · Oct 2026
- Industrycs.LG
A Stackelberg game framework lets discrete diffusion learn which token corruptions most improve the denoiser
Learning which token corruptions best train a denoiser improves discrete diffusion generation across molecules, text, and playlists.
Spotify · Oct 2026
- Top Universitycs.RO
Learned control coordinates frozen diffusion models into multi-agent generators
Coordinated multi-agent diffusion steering (CMDS) learns an amortised control to coordinate frozen pretrained diffusion models using an assembly-level reward.
UCL, TUM · Oct 2026
- Independentcs.CV
Few-step pixel diffusion outperforms 25-step teacher via banded matching and DINOv2 guidance
DMA^2 distills native-RGB diffusion into four steps, outperforming its 25-step teacher on all seven metrics.
Oct 2026
- Chinese Techcs.CV
Splitting experts by role improves video diffusion model scaling
SplitMoE improves video diffusion by separating experts into semantic and generic groups to avoid spatiotemporal fragmentation.
University of Chinese Academy of Sciences, ByteDance · Oct 2026
- Top Universitycs.LG
Data-free distillation of diffusion models with a single consistent objective
A single objective unifying sample generation and score estimation lets a student distill a frozen diffusion teacher data-free, with Wasserstein convergence and ImageNet FID 2.04 at one step.
University of British Columbia, Vector Institute for AI · Oct 2026
- Chinese Techcs.LG
Large-to-small diffusion model distillation fails due to CFG error amplification, new method resolves it
Folding a large diffusion teacher’s guided policy into a small student’s conditional branch prevents guidance error amplification and improves cross-scale distillation.
Hefei University of Technology, Tsinghua University · Oct 2026
- Top Universitycs.LGcode
Dual-expert diffusion improves long-horizon predictions of collisions and smooth dynamics
HamiFormer reduces phase-space prediction error by up to 26.3% using a dual-expert diffusion field with affine symplectic maps.
University of Science and Technology of China, Tsinghua University · Sept 2026
- Big Techcs.RO
Force-reactive control lets humanoids learn contact-rich tasks from demonstration
FLoRe lets humanoids learn contact-rich tasks by predicting hand forces alongside motion and tracking those forces while balancing.
Amazon FAR (Frontier AI & Robotics), New York University · Oct 2026
- Chinese Techcs.CV
Image generators help object detectors recognize rare categories more accurately
Tail-SD, a hybrid layout-instruction diffusion model, improves long-tail detection, achieving 4.89% rare box AP and 5.68% rare mask AP on LVIS.
University of Auckland, Dcar · Nov 2026
- Top Universitycs.CV
Revising skipped diffusion steps with later feedback improves caching efficiency and quality.
PoSTCache revises earlier diffusion predictions using corrective feedback from later computations, improving caching fidelity without training.
Tsinghua University · Oct 2026
- Chinese Techcs.CV
Single-stage RL method unifies preference alignment and few-step image generation
AdvFidelity aligns human preferences and generates high-fidelity images in few steps using a single RL stage.
Alibaba, Shanghai Jiao Tong University · Oct 2026
- Chinese Techcs.CV
Future chunk prediction improves physical plausibility in autoregressive video generation
Adding Future Chunk Prediction as an auxiliary training objective improves physical plausibility in autoregressive video generation without changing inference.
Nanjing University, ARC Lab, Tencent IEG · Oct 2026
- Top Universitycs.LG
Diffusion reward models capture multimodal preference that scalar scores miss
DRM models reward as a multimodal conditional density, matching or surpassing baseline reward models.
Tsinghua University, The Chinese University of Hong Kong · Oct 2026
- Chinese Techcs.CV
One-step diffusion framework restores 1080p video in real time with efficient streaming attention
FastVR restores high-resolution video at 11 FPS using a one-step diffusion model with a lightweight VAE and chunk-wise causal attention.
Alibaba Group, Xidian University · Oct 2026
- Big Techcs.CL
Iterative draft editing speeds up speculative language model decoding
DEdit improves speculative decoding speed by iteratively editing drafts with bidirectional context, achieving over 5.7x speedup on Qwen3 models.
University of California, Riverside, Amazon Web Services · Oct 2026
- Chinese Techcs.CV
A method that generates consistent video observations across multiple vehicles in a shared driving scene
CoDrive jointly generates world-consistent multi-vehicle driving videos with precise camera trajectory control.
Tsinghua University, Tencent Hunyuan · Oct 2026
- Chinese Techcs.LG
Three diffusion RL methods unified as divergence-constrained reward maximizers
DiffusionRFT uses exact sparsemax projection to converge faster, train more stably, and achieve top performance in regression-based diffusion RL.
Tencent · Oct 2026
- Chinese Techcs.CV
Inner pipeline safety checks match outer guardrails with fewer false positives
Safety checks inside the T2I pipeline, using the model's own embeddings and latent states, can match or exceed outer guardrails with less disruption and cost.
Alibaba AAIG · Sept 2026
- Chinese Tech
Dynamic optimization loop aligns 3D generators with 2D diffusion priors to improve realism
OREO enhances 3D generation fidelity by editing rendered views and distilling the gap back into the generator via latent contrastive learning.
The Hong Kong Polytechnic University, Tencent ARC Lab · Oct 2026
- Research Labcs.CV
Image generators benefit from naming subjects, not elaborate written plans
A think block in image generation models acts as a specification of what to draw, not as a plan that is carried out.
RIKEN AIP, South China University of Technology · Nov 2026
- Academicstat.ML
Generative models sample more efficiently when paths and steps align
Elliptic schedules paired with Zolotarev discretization achieve minimax optimal sampling for Gaussians and strong log-concave guarantees.
Laboratory for Information and Decision Systems (LIDS), Massachusetts Institute of Technology · Nov 2026
- Big Techcs.RO
Robots can improve actions by changing randomness used for video prediction
Low-Rank Noise Steering (LoRN) improves robot policies by steering video noise via a low-rank projection, boosting success by 26.2 percentage points.
NVIDIA · Nov 2026
- Top Universitycs.CV
Speculative decoding speeds up vision models by 3.58 times with visual-token weighting
MMFlash speeds up vision-language model inference by up to 3.58x using adaptive weight adjustments for visually dependent tokens.
National University of Singapore · Nov 2026
- Chinese Techcs.AI
AI video generator uses separate visual and audio timelines to control cinematic dialogue
Découpage independently controls audio and visual timing for long multi-shot video generation, enabling cinematic coherence.
The Hong Kong University of Science and Technology, Ant Group · Nov 2026
- Chinese Techcs.CV
Video generators can both follow human movement and recover it
EgoGeneralist adapts pretrained video generators to unify egocentric action-conditioned video generation, pose estimation, and joint action-video generation.
Monash University, Ant Digital Technologies, Ant Group · Nov 2026
- Chinese Techcs.CV
Motion model understands and generates in one shared space
Möbius unifies motion understanding and generation by diffusing noisy motion directly in the same multimodal space as text.
The Hong Kong University of Science and Technology, Tencent Games Motus · Nov 2026
- Academiccs.CV
Tuning only the first denoising step restores lost diversity in distilled models
Diversity loss in distilled diffusion models is concentrated in the first step and can be recovered by a lightweight first-step bias adapter.
Graz University of Technology, Complexity Science Hub Austria · Oct 2026
- Chinese Techcs.LG
Biased training targets help diffusion models learn stronger denoisers
Swift trains a diffusion model against targets extrapolated away from a frozen weak denoiser, improving generation quality without adding inference-time cost.
University of Michigan, Ann Arbor, Alibaba Group · Oct 2026
- Chinese Techcs.RO
Visual reward optimization can improve robots while reinforcing wrong-object mistakes
Optimizing a learned visual reward can raise task success and wrong-object failures simultaneously, hiding the error.
Baidu, University of California, Santa Barbara · Oct 2026
- Big Techcs.CV
Preserving camera-motion information helps video generators maintain more consistent geometry
CausalAlign improves ego-motion consistency in video generation by aligning representations with a 3D foundation model.
Peking University, Beijing Institute for General Artificial Intelligence · Oct 2026
- Independentcs.LG
Fixed-cost drafter speeds speculative decoding as context lengths grow
LongSpark decouples drafter cost from prefix length by extracting fixed-size, multiscale views from target verification.
Oct 2026
- Big Techcs.CV
Models can isolate restricted image concepts during pretraining
Unsafe concept knowledge in text-to-image models can be isolated during pretraining into a detachable branch, preserving safe-data quality while suppressing restricted generation.
University of Wisconsin–Madison, Adobe Research · Oct 2026
- Big Techcs.CV
One LiDAR model generates, forecasts and resimulates driving scenes in real time
A single conditional LiDAR world model unifies generation, forecasting, and resimulation, real-time at 11 FPS on one H100 GPU.
NVIDIA, University of Illinois, Urbana-Champaign · Oct 2026
- Chinese Techcs.CV
ReVision adapts frozen image generators to perception tasks with per-instance parameter updates
ReVision generates instance-specific low-rank updates from instructions to adapt frozen image generators for dense perception tasks.
Tencent Hy, Shanghai University of Finance and Economics · Oct 2026
- Top Universitycs.CV
Single generator reconstructs 3D scenes at any scale without per-scene optimization
T3lescope reconstructs arbitrary-resolution high-fidelity 3D scene meshes from posed multi-view images without per-scene optimization.
Preferred Networks, Inc., The University of Tokyo · Oct 2026
- Big Techcs.CV
SemCam turns subject-relative camera instructions into adaptive video motion
SemCam generates video with a camera behavior described by a discrete label relative to a moving subject, without an explicit trajectory.
Amazon Prime Video, Tel Aviv University · Oct 2026
- Big Techcs.CV
Persistent memory keeps generated 4D worlds consistent across camera paths
Demiurge constructs navigable 4D worlds from a single video by autoregressively rerendering novel views and merging them into a dynamic 4D point-cloud memory.
USTC, CASIA · Oct 2026
- Chinese Techcs.CV
Image-based visual examples enable powerful video editing without large paired video datasets
Video editing can be achieved by routing edits through the image domain with an in-context visual demonstration, reducing the need for expensive paired video data.
National University of Singapore, ByteDance Seed · Oct 2026
- Big Techcs.LG
Keeping token uncertainty improves discrete diffusion generation and fast sampling
Simplex Diffusion Models represent beliefs over categories on the probability simplex, mitigating information collapse and achieving competitive performance on text and code generation.
Google DeepMind, EPFL · Oct 2026
- Big Techcs.CV
Canvas instructions enable precise image editing without separate text prompts.
Image editing can be directed by drawing marks and notes on the image; VibeEdit outperforms text-prompted baselines on targeted edits among similar objects.
University of Sydney, Microsoft Research · Oct 2026
- Top Universitycs.CV
Routing 3D geometry into diffusion improves sparse-view scene synthesis
VGGT-Diff achieves state-of-the-art novel view synthesis from sparse views by routing visual geometry latents from VGGT-Ω into a pretrained video diffusion model.
XPeng Motors, The Chinese University of Hong Kong · Oct 2026
- Top Universitycs.CV
Purifying alignment targets improves diffusion generation, especially for reconstructive encoders
Aligning to the residual after removing locally predictable content (LAP) improves diffusion generation quality and turns MAE into an effective alignment target.
Cornell University, University of Pennsylvania · Oct 2026
- Big Techcs.CV
More Egocentric Video Improves Agents but Not Their Object Effects
Scaling ego-centric video to 30,000 hours models agents well but leaves object interaction fidelity far behind, requiring training changes, not just more data.
University of Illinois Urbana-Champaign, Carnegie Mellon University · Oct 2026
- Top Universitycs.LG
Generative models as proposal distributions improve long-horizon decision making
Generative Trajectory Planning uses a generative model as a proposal distribution for action sequences enabling efficient long-horizon planning.
Harvard University · Oct 2026
- Big Techcs.CV
Decoupling abstract state from visual state improves video world model consistency
StateCraft pairs a language model with a video diffusion model to maintain a persistent abstract state for consistent and controllable video world simulation.
INSAIT / Sofia University, ETH Zurich · Oct 2026
- Big Techcs.CV
Frequency-aware forecasting cuts diffusion transformer evaluations while preserving fidelity
SpecKCast accelerates Diffusion Transformers by forecasting DCT coefficients with Kalman filters along noise-schedule-aligned time, achieving up to 6x speedup with minimal memory.
University of Toronto, Vector Institute · Oct 2026
- Top Universitycs.AI
Verifiable visual rewards let image generators train on synthetic scenes and transfer to natural prompts
Programmatically verifiable visual rewards, trained with reinforcement learning, transfer from synthetic geometric scenes to natural image prompts, improving instruction following.
University of Washington · Oct 2026
- Chinese Techcs.CV
Distilling geometry attention into video diffusion models improves long-range consistency
Distilling geometric correspondences from a frozen geometry model into video attention improves long-range scene and dynamic consistency without adding inference modules.
Tsinghua University, AMAP, Alibaba Group · Oct 2026
- Independentcs.CV
Native 4K image editing works better as an agent system than a single model
BluePixel combines an efficiently adapted native-4K editor with specialist routing and reward-based retries, delivering four-step 4K edits while ranking second on VINS-4KEval++.
Oct 2026
- Big Techcs.LG
Learning powered conditional distributions directly with importance-weighted flow matching
Importance-Weighted Conditional Flow Matching learns an ω-powered conditional distribution more faithfully than classifier-free guidance.
University of Geneva, HES-SO/HEG Genève · Oct 2026
- Chinese Techcs.CV
Long skip connections help diffusion transformers learn cleaner visual representations
Diffusion Transformers without long skip connections produce weaker visual features because diffusion noise pollutes the main path; adding long skip connections restores feature quality.
Institute of Information Engineering, CAS, University of Chinese Academy of Sciences · Oct 2026
- Top Universitycs.AI
Directional decomposition reveals what drives Fréchet scores across generative models
Directional Fréchet distance reveals that a few interpretable feature directions often dominate FID, FVD, and Protein FID discrepancies hidden by scalar scores.
Agency for Defense Development, South Korea, Harvard University · Oct 2026
- Big Techcs.CV
ALIVE inserts objects that respond coherently to video actions
ALIVE inserts objects into videos that coherently interact with the source actions using an edited first frame and object-only instruction.
University of Rochester, Adobe Research · Oct 2026
- Chinese Techcs.CV
Dual-frame interleaving lets holistic 3D priors produce part-decomposed assets
StructParts reuses a holistic 3D prior via dual-frame coordinate interleaving to generate part-decomposed assets with minimal finetuning.
The University of Hong Kong, Transengram · Oct 2026
- Chinese Techcs.CV
Geometry-aware alignment keeps diffusion models near natural image structure
Anisotropic geometry-aware optimization prevents off-manifold drift in diffusion model preference alignment.
Sun Yat-Sen University, Alibaba Group · Oct 2026
- Top Universitycs.CV
Large-scale dataset and token-fusion architecture improve 3D asset editing
Alchemy3D achieves better 3D asset editing fidelity, source preservation, and visual quality than prior methods.
The University of Hong Kong, Shenzhen Loop Area Institute · Oct 2026
- Chinese Techcs.CV
Benchmark exposes image generators’ limits on dense bilingual visual text
UltraText Bench exposes distinct fidelity, clarity, spatial, and scene-quality failures in dense English and Chinese text generation as workloads increase.
Westlake University, Ant Group · Oct 2026
- Academiccs.RO
Depth supervision boosts RGB and geometric consistency in robot world models
DepthWorld jointly predicts RGB and depth, improving RGB PSNR by 1.48 dB over an RGB-only baseline while providing metric depth.
Czech Technical University in Prague · Oct 2026
- Big Techcs.RO
Offline stitching of robot demos boosts policy success by 21 percentage points
Offline algorithm NEEDLE stitches verified action bridges between robot observations to improve policy success, raising real-robot success by 21 points on average.
Stanford University, NVIDIA · Oct 2026
- AI Startupcs.CV
Spatial resolution hierarchy enables parallel mesh generation with explicit topology
MeshOctave generates native meshes with parallel discrete split-and-rewire tokens across grid-resolution scales, beating autoregressive and flow-matching baselines in fidelity and validity.
Huazhong University of Science and Technology, Meshy AI · Oct 2026
- Top Universitycs.RO
Robots Can Revise Visual Plans Instead of Starting Over
Revisable Temporal Planning revises a robot's visual plan after execution feedback by continuing from a saved intermediate generation state.
Beihang University, Peking University · Sept 2026
- Big Techcs.RO
Rolling predictions make robot replanning 4.5 times faster
Rolling-WAM distributes video-action denoising across replanning cycles to achieve 4.5x faster replanning without sacrificing manipulation performance.
University of Southern California, Brown University · Sept 2026
- Chinese Techcs.CV
Fixing a training mismatch yields better video generation faster
BlockAlign enables interface-consistent preference optimization for trajectory-distilled video generators.
Alibaba Group · Nov 2026
- Big Techcs.CV
AI agents benefit from checking both plans and imagined outcomes
Tiresias plans actions by generating candidate trajectories, ranking them, then imagining their visual consequences via a world model.
University of Washington, National University of Singapore · Nov 2026
- Independentcs.LG
Reward-trained image models can get worse the more candidates you generate
Visual RL post-training can degrade the high-quality outputs that Best-of-N sampling relies on.
Nov 2026
- Independentcs.RO
World model simulator works across 22 robot platforms, enables policy learning
PrimeSim turns a pretrained video generator into an action-conditioned, multiview world model that simulates robot manipulation across embodiments at 10.1 FPS.
Nov 2026
- Big Techcs.CV
AI image generators learn to follow dense prompts by using their own visual understanding as a teacher
Structural Self-Teaching improves compositional image generation in unified multimodal models using the understanding path as an internal teacher.
Samsung Electronics, Seoul National University · Nov 2026
- Chinese Techcs.CV
Image editor handles 3D pose and identity together from one instruction
DecoEdit performs unified 3D-aware image editing by decoupling geometry from detail preservation, surpassing proprietary models.
Zhejiang University, Alibaba Group · Nov 2026
- Big Techcs.RO
Humanoid robots trained in simulation can navigate homes without prior mapping
A humanoid robot trained only in simulation achieves 91% success performing physical tasks in unfamiliar real homes.
Apple · Nov 2026
- Academiccs.RO
AI agents plan better by separating subgoals from precise actions
ID-WM decouples reachable state transitions from precise actions to plan long-horizon tasks without task-specific knowledge.
New York University, Meta-FAIR · Nov 2026
- Big Techcs.CV
Proper coordinate alignment reshuffles novel-view AI rankings
Benchmark rankings were distorted by misaligned coordinate frames; once fixed, explicit 3D Gaussian methods lead.
University of British Columbia, Google · Oct 2026
- Research Labcs.RO
Robots handle visual changes better by reusing previously successful movements
R2A makes VLA robot policies robust to visual changes by retrieving geometric motion primitives from training experience.
VinUniversity, VinRobotics · Oct 2026
- Top Universitycs.CV
Video generators better preserve subjects’ appearance as cameras circle them
OmniShowcase generates subject-consistent videos from multi-view images using geometry-aware conditioning and automated data synthesis, achieving state-of-the-art identity preservation and 3D consistency.
FLIPTION, KAIST AI · Oct 2026
- Big Techcs.RO
Hand-held skill interface enables portable data collection for dexterous robot hands
DexPort, a portable skill-interface for dexterous hands, boosts task success rates from 11% to 64% over state-only baselines.
Amazon FAR (Frontier AI & Robotics), Stanford University · Oct 2026
- Research Labcs.CV
AI image detectors can be fooled by editing tricks found online
Web-sourced image processing tools achieve up to 100% attack success against AI image detectors.
CISPA Helmholtz Center for Information Security · Oct 2026
- Big Techcs.CV
Robust prediction principle yields better latent spaces for diffusion models
Recovering semantic features and clean images from corrupted latents makes visual tokenizers easier for diffusion models to learn without sacrificing reconstruction fidelity.
Tsinghua University, Microsoft Research · Oct 2026
- Big Techcs.CV
Fixed auxiliary cameras make multi-agent world models more consistent
Adding fixed auxiliary camera views to a multi-agent video world model improves consistency between agents and reduces token cost by encoding only changes from a static reference.
Meta AI, University of Oxford · Oct 2026
- Chinese Techcs.CV
Video generators preserve existing text better than they edit it
Video text generation capability is non-monolithic: I2V sustains text best, V2V editing is weakest, and degradation concentrates on structural and temporal factors.
Alibaba Group, Shanghai Jiao Tong University · Oct 2026
- Chinese Techcs.CV
Geometry-aware test-time scaling improves camera control in video diffusion without retraining
Training-free geometry-aware candidate selection at test time improves camera control and geometric consistency in video diffusion models without retraining.
University of Freiburg, Huawei Heisenberg Research Center (Munich) · Oct 2026
- Big Techcs.LG
Few-step diffusion can sacrifice output coverage for single-sample quality
Few-step diffusion distillation can improve single-sample quality while narrowing valid-output coverage, especially when trained with distribution-matching objectives.
Rice University, Apple · Oct 2026
- Chinese Techcs.CV
Projecting critic errors stabilizes four-step video diffusion distillation
Projected Distribution Matching Distillation filters critic errors in video diffusion model distillation with a one-line code change.
ByteDance Inc., University of California, San Diego · Oct 2026
- Top Universityq-bio.BM
Fold’EM infers atomic structures directly from cryo-EM particle images
Fold'EM infers atomic structures directly from cryo-EM particle images by combining protein structure priors with particle evidence, bypassing density reconstruction.
Institute of Science and Technology Austria, University of Zurich · Oct 2026
- Top University
Dynamics-aware distillation preserves motion in four-step video world models
DyMD improves few-step video generation by preserving interaction dynamics via distribution matching distillation.
Beihang University, JD Future Academy · Sept 2026
- Research Labcs.CV
Image generators improve clock accuracy from under 1% to over 60%
Geometric token encoding with periodic features enables precise angular control in text-to-image generation.
Bilkent University, CISPA Helmholtz Center for Information Security · Nov 2026
- Chinese Techcs.AI
Image generators follow object relationships more faithfully with targeted prompt edits
StruPO uses calibrated scene graphs to learn prompt edits that improve image quality without changing the requested scene.
Zhejiang University, Tencent AI Lab · Oct 2026
- Top Universitycs.RO
Unlabelled navigation videos can teach robots how scenes change as they move
OpenNWM learns universal latent actions from action-free video to generalize navigation world models across environments.
CASIA, University of Chinese Academy of Sciences · Nov 2026
- Chinese Techcs.LG
AI image generators and robot controllers can follow guidance more reliably
Coupled Guidance preserves the latent-data pairing during flow-matching generation, improving stability and sample quality in robotics and image tasks without extra overhead.
Huawei · Oct 2026
- Independentcs.CVcode
Image generators can outperform their multi-step originals on image statistics
ROAD achieves state-of-the-art one-step text-to-image generation using only a single adversarial objective and a lightweight critic.
Oct 2026
- Top Universitycs.LG
A framework predicts RL post-training outcomes without running reinforcement learning
PoEM predicts reinforcement learning outcomes for new reward functions without retraining by composing existing policies.
MIT CSAIL · Sept 2026
- Top Universitycs.RO
Touch-directed curiosity helps robots discover useful manipulation skills
Tactile-driven curiosity lets robots learn manipulation and grasp policies without task rewards or demonstrations, and improves post-training of VLA models.
ETH Zürich, University of California, Berkeley · Oct 2026
- Chinese Techcs.AI
Fast generative models improve with RL when noise split is optimized geometrically
Geo-FSRL selects stochastic policies minimizing conditional KL divergence, achieving higher reward scores than prior methods in few-step flow models.
Beihang University, Sensetime Research · Nov 2026
- Research Labcs.CV
Vision inputs expose greater social bias than text in matched tests
Switching from text to vision reduces fairness in multimodal LLMs by 0.291, a gap not due to poor visual comprehension.
Technical University of Munich, Helmholtz Munich · Oct 2026
- Chinese Techcs.CV
AI model controls characters in unfamiliar game worlds in real time
An agentic game world model that generalizes zero-shot to novel scenes, characters, and motions with real-time interaction.
Tencent, National University of Singapore · Oct 2026
- Chinese Techcs.CV
AI video generators reduce flickering and preserve detail with few steps
Adding a predictive representation loss to distribution matching distillation improves few-step video generation spatial fidelity and temporal coherence.
Shanghai Jiao Tong University, Hunyuan (Tencent) · Oct 2026
- Top Universitycs.RO
Large-scale benchmark and end-to-end model achieve robust dexterous grasping in cluttered scenes
A 2.6-million-scene benchmark and end-to-end model achieve robust dexterous grasping in cluttered scenes.
ACE Robotics, CUHK, MMLab · Oct 2026
- Chinese Techcs.LG
Distribution matching distillation and drifting models are equivalent under finite-sample analysis
Distribution Matching Distillation is equivalent to Drifting Models, and multi-bandwidth training improves DMD convergence and quality.
Baidu Inc., Central South University · Oct 2026
- Top Universitycs.CV
Shared 3D controls coordinate camera and object motion in generated video
Generative Cinematographer (GenCine) lifts a single image to 3D to jointly author camera and foreground motion using local 3D motion handles.
Johns Hopkins University · Oct 2026
- Chinese Techcs.CV
Latent trajectory guidance improves video generation reasoning
VR-JEPA uses latent trajectory prediction to guide video generation for improved visual reasoning.
Sun Yat-sen University, Shenzhen Loop Area Institute · Oct 2026
- Academiccs.AI
Concept entanglement forces a fundamental trade-off in diffusion model unlearning
Concept unlearning in diffusion models is fundamentally limited by the geometric overlap of target and neighbor concepts in activation space.
University of Illinois Urbana-Champaign · Oct 2026
- Research Labcs.LG
AI models excel at cyclone track prediction but struggle with intensity and uncertainty
Leading AI weather models outperform numerical forecasts on cyclone tracking but retain intensity biases and calibration errors.
INRIA, AXA · Oct 2026
- Big Techcs.CV
World model learns where to predict and where to generate
D3WM jointly predicts deterministic next states and spatial uncertainty maps to guide diffusion for consistent future frames.
University of Michigan, NVIDIA · Oct 2026
- Chinese Techcs.CV
LightVAE speeds up video autoencoders by exploiting temporal and depth redundancy
LightVAE accelerates video VAE decoding by up to 2.24x through temporal low rank compression and block pruning.
SenseTime, Beihang University · Oct 2026
- Chinese Techcs.CV
TempoKV compresses video generation KV caches by exploiting key-value predictability asymmetry
TempoKV predicts keys from history and values from current keys to compress KV cache up to 6.71x for video generation.
Shanghai Jiao Tong University, South China University of Technology · Oct 2026
- Chinese Techcs.CV
Dense matching can track identity beyond physical continuity
FreeMatching learns identity-preserving dense correspondence across image editing and generation tasks by combining foundation representations, heterogeneous supervision, and teacher-guided refinement.
The University of Hong Kong, ByteDance Seed · Oct 2026
- Independentcs.CV
Robot policies generalize better when future tokens remain at inference
Latent world action models match explicit ones in distribution but lose generalization; conditioning on fully noised future tokens recovers most of the benefit.
Oct 2026
- Top Universitycs.RO
Guided by failures, choosing initial states improves robot learning per supervised episode
Mulligan selects initial states from failures to improve robot learning faster than uniform collection.
Stanford University · Oct 2026
- Chinese Techcs.CV
Decomposing video editing into keyframe editing and motion-guided interpolation
Video editing is achieved by editing sparse keyframes with an image editor and interpolating with motion-guided video diffusion.
ByteDance Seed, Johns Hopkins University · Oct 2026
- Chinese Techcs.CV
Object-centric 3D memory enables consistent video generation under large viewpoint changes
Totem uses object-centric 3D memories to generate videos with consistent objects across large viewpoint changes.
Harbin Institute of Technology, Huawei Technologies Ltd. · Oct 2026
- Chinese Techcs.CV
Multi-coordinate conditioning improves camera control in generated video
Cross-coordinate conditioning fusion enables explicit control of camera motion, object motion, and depth-of-field in video generation.
Tencent Video Solar System · Oct 2026
- Big Techcs.CV
Polynomial student paths improve four-step video distillation quality
Parametric Trajectory Distillation fits teacher trajectory segments to polynomials and supervises along the student's own predicted path, preserving motion and detail in four-step video generation.
EPFL, NVIDIA · Oct 2026
- Big Techcs.RO
Synthetic views make robot policies more tolerant of pose errors
MobileVISTA generates pose-perturbed training data from single-pose demonstrations to improve mobile manipulation policy robustness.
Stanford University, Toyota Research Institute · Oct 2026
- Big Techcs.AI
Local counterfactual edits outperform global unsafe subspace removal
Text-to-image safety fails when using a single global unsafe subspace due to a coverage-selectivity trade-off.
KAIST, Samsung Electronics · Oct 2026
- Big Techcs.RO
Flexible block-causal models sustain efficient long-horizon action-conditioned rollouts
A block-causal world-action model enables real-time, stable long-horizon imagination and planning, solving benchmark tasks entirely in imagination.
Center for Robotics and Embodied Intelligence (CREO), New York University, Courant Institute of Mathematical Sciences · Oct 2026
- Top Universitycs.LG
Loop-specific targets let diffusion transformer improve with extra inference loops
LiFT improves ImageNet generation by repeatedly applying a shared DiT core with depth-specific trajectory targets, cutting parameters and compute versus a dense DiT baseline.
University of Amsterdam, TNO · Oct 2026
- Top Universitycs.RO
Robot learns manipulation tasks from scratch in minutes with AI-generated 3D models
ZeroBot enables robot manipulation learning from scratch in minutes using image-to-3D generative models and parallel RL.
Imperial College London, Robotics and AI Institute · Oct 2026
- Top Universitystat.ML
Steepest Guidance aligns generative models by maximizing local reward improvement
Inference-time alignment of flow and diffusion models can be cast as sequential optimization in probability-measure space, where steepest local ascent of the reward functional gives unbiased guidance.
LY Corporation, The University of Tokyo · Oct 2026
- Chinese Techcs.CV
Diagnosing where video models collapse diversity across dimensions
Video generation models collapse diversity on specific dimensions like Motion and Camera, even when global scores are high.
Renmin University of China, Alibaba Group · Oct 2026
- Chinese Techcs.CV
Velocity corrections improve instruction following in few-step text-to-image generation
A sub-20M-parameter velocity residual corrector improves few-step text-to-image instruction following by 17.60 percentage points without altering the frozen generator or adding forward passes.
The University of Osaka, Alibaba Group · Oct 2026
- Big Tech
Trajectory optimization that preserves endpoints improves closed-loop driving
Endpoint-constrained optimization repairs intermediate waypoints while preserving predicted endpoints, substantially improving closed-loop performance across six end-to-end driving policies in two simulators.
University of Toronto, Vector Institute · Oct 2026
- Research Labcs.DC
Adaptive verification speeds block-diffusion decoding under heavy concurrent demand
DScale improves throughput of block-diffusion speculative decoding by up to 48.8% over DFlash with only a 112K-parameter predictor.
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, University of Chinese Academy of Sciences · Oct 2026
- Chinese Techcs.GR
Benchmark shows image editing models fail to parse scribble-only intent
Existing VLM/LLM-based image editing models fail to interpret scribble-only editing intents; a soft-token baseline improves understanding.
MIT, ByteDance · Oct 2026
- Research Labcs.CV
Backdoor attacks on interactive video generation make outputs freeze when action triggers appear.
Backdoor attacks on interactive video generation models can use action-guided triggers to cause frozen frames that ignore user actions.
Chinese Academy of Sciences, University of Chinese Academy of Sciences · Oct 2026
- Top Universitycs.CV
Separating context-writing and denoising parameters improves autoregressive video generation
Separating parameters for context writing and denoising in autoregressive video generation improves visual quality and long-horizon consistency without extra data or fine-tuning.
Tsinghua University, JD · Oct 2026
- Big Techcs.AI
SatisDive method navigates the satisfaction-diversity frontier for text-to-image generation
Text-to-image generation can satisfy user preferences while maintaining diversity by ensuring each image meets a minimum reward threshold.
University of Central Florida, Amazon · Oct 2026
- Chinese Techcs.CV
One-step action generation and asynchronous pipeline achieve real-time world action models
RealtimeWAM enables near-lossless real-time action generation with 25x speedup via one-step denoising and asynchronous pipelining.
Nanyang Technological University, Beihang University · Oct 2026
- Academiccs.CV
Video RL improves by assigning credit to individual visual tokens
TVRL uses gradients from a frozen vision-language model to allocate token-level credit, outperforming scalar-reward GRPO on video generation.
Northeastern University · Oct 2026
- Chinese Techcs.CV
Explicit weather transitions improve control in generated future-scene videos
MeteoVerse generates future videos from a single image with explicit control over weather transitions including introduction, preservation, and removal.
Harbin Institute of Technology, Huawei · Oct 2026
- AI Startupcs.CV
Complete 3D meshes improve control of generated video motion
Using explicit rigid 3D geometry to control video world models achieves better camera and object control than prior methods.
Stability AI, University of Illinois Urbana-Champaign · Oct 2026
- Big Techcs.CV
One-step video refiner achieves 8.91x speedup for 4K upscaling
SoL-Refiner transforms low-resolution video outputs into 4K videos with a single denoising step, outperforming multi-step refiners.
NVIDIA · Oct 2026
- Chinese Techcs.CV
Independent chunk scoring reduces drift in long autoregressive video rollouts
Rollout-Marginal Distillation scores each generated chunk independently against a teacher, then refines temporal coherence, preventing error accumulation in long autoregressive video generation.
MMLab, The Chinese University of Hong Kong · Oct 2026
- Chinese Techcs.CV
Distribution drifting on spatial features enables efficient one-step super-resolution
DriftSR achieves one-step real-world super-resolution by drifting pretrained diffusion features without additional trainable components.
Nanjing University of Science and Technology, Huawei · Oct 2026
- Big Techcs.LG
Denoiser Jacobian spectrum links to generative model quality and enables targeted improvements
Better generative denoisers have larger principal Jacobian eigenvalues, and regularizing those responses while suppressing noisy directions improves generation quality.
Stony Brook University, Microsoft Research · Oct 2026
- Academiccs.CV
A zero-shot pipeline that turns monocular 360° video generators into stereoscopic ones
A zero-shot pipeline generates 4K stereoscopic 360° video from monocular panoramic video by enforcing epipolar consistency.
UNC Chapel Hill, Dolby Laboratories · Oct 2026
- Industrycs.AI
Moving softmax to matrix operations speeds up AI inference by over 2x
Fused-Softmax reformulates softmax into matrix operations to eliminate the vector-unit bottleneck in quantized attention.
AMD · Nov 2026
- Academiccs.CV
Noise-rejuvenated SMC enables test-time scaling for flow matching image generation
Noise-rejuvenated sequential Monte Carlo enables reward-guided text-to-image generation with flow matching models, without training or gradient optimization.
Uppsala University, Linköping University · Oct 2026
- Chinese Techcs.CV
World coordinates replace learned features for game model control
ActionRoPE represents action-driven camera displacement as world-coordinate rotary positions, enabling spatially consistent control in planar-camera game world models.
Tencent, National University of Singapore · Oct 2026
- Industrycs.CV
Single-pass GAN matches diffusion quality for real-time talking heads at 43 ms latency
A single-pass GAN with acausal noise shaping matches diffusion quality for real-time audio-driven facial animation at 43 ms latency.
Meta Reality Labs · Oct 2026
- Chinese Techcs.CV
Generative framework converts sparse panoramas into annotated multi-room 3D scenes
Sparse real-estate panoramas can be turned into annotated multi-room 3D Gaussian Splatting scenes that improve VLM spatial reasoning.
The Hong Kong University of Science and Technology, ShanghaiTech University · Oct 2026
- Chinese Techcs.CV
Benchmark exposes distinct weaknesses in multimodal audio-video generation systems
ORAV is a benchmark for audio-video generation from multimodal contexts that reveals distinct model strengths and failures.
Tsinghua University, Tencent Hy · Oct 2026
- Top Universitycs.CV
GeoVerse blends geometric latent diffusion with video generative priors for consistent novel views
GeoVerse achieves world-consistent novel view synthesis by combining geometric latent diffusion with video generative priors and a global spatial memory.
Shanghai Jiao Tong University, Shanghai Artificial Intelligence Laboratory · Oct 2026
Concepts are extracted from each paper and reused across the corpus, so this page grows on its own as the desk reads.