Advancing coding in low-resource programming languages
Official research discussion of improving coding capabilities in languages with less available training material.
THE RESEARCH ATLAS
Efficient AI training papers
Efficient attention, low-rank adaptation, sparse experts and inference optimization. Read compute, memory and task quality together.
Scaling Misuse Detection with Self-Assessment
Models assess their own conversations as an additional misuse-detection layer, reusing cached context to limit overhead.
PointWalker: Mapless Quadruped Locomotion Directly From Raw LiDAR Point Clouds
Fuses LiDAR point clouds and proprioception through attention, connecting terrain perception with a reinforcement-learned quadruped policy.
DeepSeek-V4.1-Flash: smarter, faster, more efficient
Official release introducing an asymmetric causal encoder-decoder architecture and linking its technical report and model resources.
Qwen3-Coder-Next Technical Report
An open-weight model for coding agents explores strong training recipes with a small active parameter footprint.
Qwen3-ASR Technical Report
A speech-recognition family and a forced-alignment model address multilingual transcription and timestamp prediction.
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
Sparse attention and post-training methods are combined to improve efficiency, reasoning and agent-oriented behavior.
gpt-oss-120b & gpt-oss-20b Model Card
Open-weight reasoning models are described through their architecture, training, capabilities and safety evaluations.
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
A compact vision-language-action model explores accessible robot learning and an efficient inference pipeline.
Qwen3 Technical Report
Dense and mixture-of-experts models integrate thinking and non-thinking behavior, with configurable inference-time reasoning.
DeepSeek-V3 Technical Report
A mixture-of-experts language model combines architectural and training-efficiency techniques, including latent attention and multi-token prediction.
Mixtral of Experts
Introduces a sparse mixture-of-experts model exploring model capacity and activated computation.
Direct Preference Optimization: Your Language Model is Secretly a Reward Model
A preference-based objective optimizes a language model directly, avoiding a separate reward-model training stage.
FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
An IO-aware attention algorithm reduces memory transfers while computing exact attention.
Training Compute-Optimal Large Language Models
The study examines how to allocate a training-compute budget between model size and training data.
LoRA: Low-Rank Adaptation of Large Language Models
Low-rank trainable updates adapt a frozen pretrained model without updating all of its parameters.
Institutions working on this topic
Counts are records from that institution that carry this topic in this index.