HuggingFace
Continual Learning in Transition
Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through par...
🤖 AI 前沿资讯
每天自动聚合 AI 领域最新动态
HuggingFace
DRIFT: Derailing Denoising Trajectories of Flow-Matching VLAs with Adversarial Patch Attack
Flow-matching vision-language-action (VLA) models such as pi0 generate robot actions by integrating a learned denoisi...
HuggingFace
Resume Means Resume: A Machine-Checked Conformance Contract for Checkpoint, Interrupt, and Resume Semantics in Workflow Persistence Layers
A framework that persists execution state so a run can be interrupted, survive a crash, and continue must decide what...
HuggingFace
Lossless Tensor Compression as Program Synthesis
Model checkpoints are growing in both number and size, which makes archival, transfer, and deployment increasingly co...
HuggingFace
SIGNPOST-Bench: Benchmarking Text-Vision Conflict Resolution in Multimodal Large Language Models
Multimodal large language models (MLLMs) make grounded predictions in real-world scenes by combining visual and textu...
HuggingFace
What AI Red-Team Evaluations Can and Cannot Prove
Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable ...
HuggingFace
FinanceHarness: Autonomous Financial Deep Research Framework
Powered by advances in LLMs and autonomous agents, deep research has become one of the most widely adopted agentic pr...
HuggingFace
Helping Music Co-Creation Agents 'Listen' Well: Hierarchical Self-Supervised World Models for Understanding and Generation
Collaborative music agents need internal representations rich enough to support both understanding and generation, ye...
arXiv
VQ-VAD: Vector-quantized Motion Representation Learning for Human-centric Video Anomaly Detection
Video Anomaly Detection (VAD) is inherently challenging due to the scarcity of anomalies and the large visual variabi...
arXiv
MultiPathFormer: Towards a Foundation Model for Multipath Wireless Propagation
Recent advances in machine learning have enabled training of wireless foundation models, which aim to support tasks s...
arXiv
Capability-Gated Planning: Cost-to-Goal Discovery and the Limits of Myopic Experiment Selection
Systems that automate scientific discovery must repeatedly decide which experiment to run, which hypothesis to test, ...
arXiv
Item Response Theory for AI Safety
Language models differ in how safely they behave and these differences are measured by safety benchmarks. But aggrega...