技术雷达 · 2026-09-30

G GitHub 新星 10 条

A arXiv 论文 10 条

1 FurE: Efficient Instance-Specific 3D Fur Reconstruction without Animal-Fur Datasets cs.CVcs.AIcs.GR
Realistic and editable animal fur reconstruction from multi-view images is challenging due to fine-scale detail, self-occlusion and obfuscation, and, unlike human hair, the lack of animal-fur datasets
Srinjay Sarkar, Prakhar Kaushik, Soumava Paul
2 Telescopic Language Models cs.CLcs.AI
One deployed language model must often serve many compute budgets, yet serving each budget still means a separate training or compression run per point. We train a Telescopic Language Model (TLM) to b
Zhilin Guo, Boqiao Zhang, Hakan Aktas
3 PDMD: Projected Distribution Matching Distillation for Video Diffusion Models cs.CVcs.LG
Modern video diffusion models require tens of denoising evaluations over long spatiotemporal token sequences. Distribution Matching Distillation (DMD) reduces the number of function evaluations (NFE)
Zimo Wang, Junkun Yuan, Angtian Wang
4 Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning cs.CVcs.AI
Unified multimodal models can both look at and render images, so in principle they can repair their own generations: diagnose what an image gets wrong, revise it, observe the result, and diagnose agai
Yijia Fan, Ziqi Huang, Zhongang Cai
5 Unifying Distributional Training for One-Step Visual Generation cs.LG
\emph{Distributional training} provides collective supervision for one-step visual generation by matching real and generated features in frozen representation spaces. We introduce \emph{a unified theo
Chi Zhang, Haoyang Shi, Yueyi Liu
6 TokenCast: Forecasting Token Consumption During LLM Agent Execution cs.LGcs.AIcs.SE
When a large language model (LLM) agent executes the same task, token consumption can vary by over an order of magnitude across runs. The agent chooses its next steps based on tool feedback and interm
Chaoqian Ouyang, Ling Yue, Libin Zheng
7 Statistical Learning of Contractive Dynamical Representations for Composite Adaptive Control eess.SYcs.LGcs.RO
We present a representation-learning framework for composite adaptive tracking control under dynamically coupled disturbances. The framework connects classical disturbance-accommodating control (DAC)
Min Kim, José Leonardo Brenes, Fred Hadaegh
8 Neural Harmonic Measure Operator cs.LGmath.NA
We introduce Neural Harmonic Measure Operator (NHMO), a neural solver for elliptic PDE problems on variable-shape domains. The harmonic measure of a domain is the boundary probability distribution tha
Jinjin He, Sinan Wang, Yuchen Sun
9 How to Loop MoE: Flatten the Experts, Untie the Attention cs.LGcs.AIcs.CL
Looped Transformers reuse one block of layers several times: by spending extra computation they push a model of fixed size further, and so use its parameters more fully; while sparse mixture-of-expert
Shouren Wang, Chuang Ma, Mohsen Hariri
10 KV-streams for Efficient Compaction in Agentic Reinforcement Learning cs.LGcs.AI
Scaling the horizon of agentic LLMs is bottlenecked by the need to fit ever longer context traces in GPU memory. Context compaction has been the most popular mechanism to alleviate this issue, keeping
Emiliano Penaloza, Dane Malenfant, Dheeraj Vattikonda

H Hacker News 10 条