Backdoored vision-language-action (VLA) policies can preserve benign task performance while producing malicious actions when a trigger appears. Detecting such activation is difficult because malicious behavior can comprise individually plausible actions, while unfamiliar tasks introduce legitimate changes in observatio...
Zi-Run Zhou, Jing-Feng Zhang, Hao-Chuan Xu et al.· 0 citations
RAIL is introduced, a human-centric benchmark grounded in the Cattell--Horn--Carroll (CHC) framework, which formalises auditory cognition into five core capabilities, and develops them into structured evaluation tasks with principled data curation and human-aligned evaluation protocols.
Hong-Yu Jin, Siyi Wang, Yang Xiao et al.· arXiv.org· 3 citations
Interactive video world models need to generate each video chunk efficiently while responding faithfully to user controls. Many systems use chunk-wise autoregressive generation with few-step denoising, but each chunk still requires several costly denoising iterations. Training-free caching can reduce this cost, yet exi...
Shang-Ye Song, Dong Gong, Hong Jia et al.· 0 citations
A common principle of effective learning is to practice material that is neither already mastered nor too difficult to permit progress. We ask how to apply this principle to structured reasoning tasks such as Sudoku and maze solving. In these tasks, a model can repeatedly revise an incomplete or incorrect candidate sol...
Hong-Bo Chen, Guo-Hua Lu, Ting Dang et al.· 0 citations
Perception-Grounded TTRL (PG-TTRL) is proposed, which aligns label-free test-time optimisation with perceptually grounded reasoning, encouraging the model to structure its reasoning more strongly on the audio input, showing the value of perceptual-grounding optimisation for test-time audio reasoning.
Jia-Heng Dong, Xiao-Feng Yu, Jean Honorio et al.· 0 citations
Flower Hub is presented, a platform for publishing, discovering, and executing decentralized and federated applications that enables reproducible benchmarking by packaging benchmarks as executable, versioned applications with standardized metadata, pinned dependencies, and explicit evaluation workflows.
Yan Gao, Mohammad Naseri, Javier Fernandez-marques et al.· 0 citations
LIRA is introduced, a local cross-layer action-conditioning mechanism that formulates VLM-to-action conditioning as depth-aware information routing and improves the principal aggregate metrics over the VLA-Adapter baseline under the same 0.5B-parameter configuration.
Zhe-Wei Zhang, Pu Wang, Guan-Ren Qiao et al.· 4 citations
MADBench is introduced, the first benchmark that treats speech and environmental audio as distinct acoustic components, enabling component-aware evaluation of audio deepfake detection across independently manipulated forgery sources and establishes a rigorous foundation for future research into robust, component-aware...
Yan-Qiu Li, Yang Xiao, Jisheng Bai et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
