HuggingFace
The emergence of "vibe coding" platforms, where users describe applications in natural language and AI agents autonomously generate full-stack software, has created a need for rigorous evaluation beyo…
HuggingFace
HuggingFace
Self-consistency detects hallucinations by generating multiple sampled answers to a question and measuring agreement, but this requires repeated decoding and can be sensitive to lexical variation. Sem…
HuggingFace
HuggingFace
Recent progress in multimodal large language models (MLLMs) has brought AI capabilities from static offline data processing to real-time streaming interaction, yet they still remain far from human-lev…
HuggingFace
HuggingFace
We present TT4D, a large-scale, high-fidelity table tennis dataset. It provides 140+ hours of reconstructed singles and doubles gameplay from monocular broadcast videos, featuring multimodal annotatio…
HuggingFace
HuggingFace
In single-stream autoregressive interfaces, the same tokens both update the model state and constitute an irreversible public commitment. This coupling creates a silence tax: additional deliberation p…
HuggingFace
HuggingFace
Distillation-based acceleration has become foundational for making autoregressive streaming video diffusion models practical, with distribution matching distillation (DMD) as the de facto choice. Exis…
HuggingFace
HuggingFace
Adapting reasoning models to new tasks during post-training with only output-level supervision stalls under reinforcement learning from verifiable rewards (RLVR) when the initial success probability p…
HuggingFace
HuggingFace
Reinforcement learning (RL) has become a central post-training tool for improving the reasoning abilities of large language models (LLMs). In these systems, the rollout, the trajectory sampled from a …
HuggingFace
HuggingFace
Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both…
HuggingFace