paperHugging Face Daily Papers·21 Oct 20264.8
FrontierCS 2 benchmarks agents on frontier computer science research tasks.
paperHugging Face Daily Papers·20 Oct 20264.4
Autonomous AI research teams discover linear attention architectures.
paperHugging Face Daily Papers·21 Oct 20264.2
Quant-Arena benchmarks LLM agents on quantitative trading research.
paperHugging Face Daily Papers·22 Oct 20264.1
Traces the Wait token in reasoning models back to pretraining data rather than RL.
paperHugging Face Daily Papers·21 Oct 20264.1
Full-stack framework for long-horizon embodied robot agents.
paperHugging Face Daily Papers·21 Oct 20264.1
LaLa-World scales JEPA latent world models with natural language actions.
paperHugging Face Daily Papers·21 Oct 20264.1
PACE encodes linear-attention recurrent states in four bits.
paperHugging Face Daily Papers·22 Oct 20264.1
Factor-aware predictive learning makes MLLM latents useful for cinematic video generation.
paperHugging Face Daily Papers·22 Oct 20264.1
RIDAR detects and intervenes on research integrity violations by autonomous research agents.
paperHugging Face Daily Papers·22 Oct 20264.0
Naive RAG beats complex agentic memory on long-term conversation tasks.
paperHugging Face Daily Papers·21 Oct 20264.0
Models future evolution to improve autoregressive chunked video generation.
paperHugging Face Daily Papers·21 Oct 20263.9
Item-level adaptive recursion spends more compute on hard items in sequential recommendation.
paperHugging Face Daily Papers·20 Oct 20263.9
Meta Expert shares experts across layers to cut MoE storage and retrieval.
paperHugging Face Daily Papers·21 Oct 20263.9
Agent Mosaic detects confidentiality leaks in agentic LLM systems.
paperHugging Face Daily Papers·21 Oct 20263.9
Agentic world action models for robot manipulation.
paperHugging Face Daily Papers·20 Oct 20263.9
Continued pretraining for embodied agents in many virtual worlds.
paperHugging Face Daily Papers·21 Oct 20263.9
Training-free output-driven allocation of block-sparse attention in video diffusion transformers.
paperHugging Face Daily Papers·20 Oct 20263.9
Qwen-Audio-3.1-ASR technical report.
paperHugging Face Daily Papers·20 Oct 20263.9
Stratified sparse attention narrows the gap to dense attention at extreme sparsity for long contexts.
paperHugging Face Daily Papers·22 Oct 20263.9
Skeleton-grounded latent action model for robot world models.
paperHugging Face Daily Papers·21 Oct 20263.8
Self-distilling future video representations for world action models.
paperHugging Face Daily Papers·20 Oct 20263.8
Diffusion LMs speed up deep research pipelines.
paperHugging Face Daily Papers·22 Oct 20263.7
IQuest-Q1 is a 320B sparse MoE model built for agentic CLI tasks.
paperHugging Face Daily Papers·21 Oct 20263.7
SplatLab automates reproducing and extending 3DGS research.
paperHugging Face Daily Papers·21 Oct 20263.7
Self-evolving harnesses for history-dependent robot manipulation.
paperHugging Face Daily Papers·21 Oct 20263.7
RealMe builds immersive volumetric video from sparse captures.
paperHugging Face Daily Papers·20 Oct 20263.6
How RL papers get into seven open-source LLM training frameworks.
paperHugging Face Daily Papers·22 Oct 20263.6
Agents create physically feasible brick builds.
paperHugging Face Daily Papers·20 Oct 20263.6
Applies coding-agent post-training to agentic robotics.
paperHugging Face Daily Papers·21 Oct 20263.6
Survey of multimodal embodied agents from computer use to robot use.