Language-conditioned robot policy that reuses a pretrained geometric foundation model and inserts a causal future predictor at an intermediate layer so the same backbone produces future 3D-aware features and action outputs, enabling geometry-aware temporal prediction with minimal architectural change.
Converts large-scale egocentric human videos into robot-format pseudo-action trajectories and introduces ACE-EGO-0, a VLA pretraining framework that unifies camera-space actions, morphology conditioning, and reliability-aware weighting to jointly learn from noisy human and high-quality robot data for improved robotic manipulation transfer.
Provides a dual-path approach for spatial vision-language models: a Language-Only Reasoning (LOR) path for stepwise linguistic deduction and a Detect-Then-Reason (DTR) path that detects 3D cues via region tokens before numerical inference. Trains with chain-of-thought cold-start supervision and reinforcement learning to improve 3D grounding and multi-step spatial reasoning.
Evaluates multimodal LLMs' ability to reconstruct past observations and act in controllable non-Markov games. Introduces RNG-Bench with two games (Matching Pairs, 3D Maze), three controllable difficulty axes, a head-to-head duel protocol, and a Memory Gap metric to separate forgetting from action errors.
Provides a rubric-based benchmark that converts dense image captions into instance-specific atomic checks (Must-Right and Easy-Wrong) and a gated scoring rule, aiming to expose perceptual brittleness and better align multimodal model evaluation with human judgment.
Adapts pretrained Vision-Language-Action (VLA) models to new camera poses and robot embodiments from a single demonstration by performing weight-vector arithmetic that injects domain-specific information. Filters noise via subspace alignment of singular components; designed for one-shot adaptation under visual and embodiment shifts.
Xingyu Zheng, Xianglong Liu +5State Key Laboratory of Complex & Critical Software Environment, Beihang University, School of Computer Science and Engineering, Beihang University +6
Accelerates text-to-image diffusion for pretrained flow-matching models using a staged low-to-high-resolution pipeline: fast low-res sampling, pixel-space GAN super-resolution, light latent noising, and short high-res refinement — >10× end-to-end speedups without retraining.
AI Video Papers2026
Generates real-time, infinite-length interactive videos of voice-controllable digital characters — 540p at up to 42 FPS on consumer GPUs. Uses TurboDiffusion and TurboServe to maintain temporal coherence without blur or drift, and accepts custom person, anime, or pet images plus selectable voice tones.
Trains a single diffusion model that unifies 3D scene reconstruction and generative modeling by operating directly in pixel/rendered-image space. Supervises diffusion on rendered views and adds a geometry-perception loss from a pretrained 3D foundation model, reducing latent information loss and improving 3D fidelity.
Expresses diverse computer-vision tasks as instruction-driven text, image, or mixed generation from a single unified multimodal model, producing outputs for detection, segmentation, depth, pose, OCR and more. Trained on a converted SenseNova‑Vision instruction–response corpus and requires no task-specific prediction heads.
AI Video Papers2026
Pretrains a DiT-based Mixture-of-Experts video foundation model for embodied intelligence by augmenting internet videos with robot-centric footage and using a multi-dimensional reward system to prioritize physical realism and task completion while scaling MoE for better capacity vs. inference trade-offs.
Reconstructs historical experience into latent memory tokens and weaves short- and long-term latent memories directly into vision-language-action reasoning to improve long-horizon robotic manipulation. Uses a four-part pipeline (curator, seeker, condenser, weaver) so memory participates natively in multimodal action formation.