AI Image2026
Generates images from natural-language prompts as an 8-step distilled checkpoint of Krea 2, optimized for fast iterative text-to-image workflows with style references and 1K–2K resolution outputs.
AI Model2026
Performs one-shot, long-horizon OCR and document parsing by using Reference Sliding Window Attention (R-SWA) to keep the decoder KV cache constant, enabling single-pass multi-page transcription; code, model weights and an accompanying arXiv report are provided.
AI Audio2026
Local English text-to-waveform TTS producing a single fixed synthetic voice in a deployable package below 10M parameters. Offers deterministic seeds, punctuation-aware long-text chunking, CPU/CUDA and ONNX runtime options, measured evaluations and a compact FP32 footprint; English-only, one voice.
AI Audio2026
Generates English speech locally from text into 24 kHz waveforms with a fixed synthetic male voice. Complete text-to-waveform TTS under ~4M parameters (≈16 MB FP32), supports CPU/CUDA inference, deterministic seeds, long-text chunking and an ONNX export path under Apache-2.0 license.
AI Model2026
Performs zero-shot classification and regression on mixed numerical and categorical tabular data by treating training rows as in-context examples and predicting in a single forward pass. Uses alternating row/column attention and row compression; limited to 10 classes and model weights are non-commercial.
AI Model2026
Diffusion-based generative model for scene and video synthesis, providing full Diffusers checkpoints and scene LoRA for fast adaptation. Includes Stage‑1 nano (1.3B) and pro (5B) variants and modular transformer/VAE components.
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 Model2026
Deployment-optimized hybrid MoE LLM (75B total / 9.3B active) produced via Iterative Puzzle compression and Multi-Token Prediction to double server throughput and raise single-GPU concurrency; designed for multilingual reasoning, long-context generation, and high-volume agentic/chat deployments.
AI Dataset2026
Provides 2,056 penetration-free cloth simulation trajectories (240 frames each, 493,440 frames, ~33 GB) across human garments, robotic manipulation, and object-collision scenarios. Includes per-vertex positions, per-frame displacements, mesh topology and collision fields under CC BY 4.0 — useful for training and evaluating learning-based cloth simulators.
AI Video2026
Generates image-to-video world-model outputs using a distilled 14B causal model optimized for chunked, KV-cached inference across long-horizon interactive scenes; offers a real-time 'causal-fast' variant capable of driving near‑real‑time video streams and an agentic harness for action-driven scene synthesis (CC BY‑NC‑SA).
Introduces KronQ, a post-training quantization framework that incorporates gradient covariance via a Kronecker‑factored Hessian to guide input/output weight rotations and sensitivity-driven mixed-precision allocation. Demonstrates stable 2-bit weight-only quantization on LLaMA-3-70B (7.93 PPL).