A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Unknown
FreeNeRF introduces free frequency regularization to improve few-shot neural rendering by addressing frequency-related challenges in NeRF training.
Unknown
Q-dit introduces a post-training quantization method for diffusion transformers that reduces memory and latency while maintaining generation quality.
Unknown
A large-scale measurement of benchmark data contamination across 17 frontier language models, quantifying how evaluation data leaks into training corpora.
Unknown
Proposes stratified per-question probability evaluation and step-wise mitigation to detect and reduce benchmark contamination in LLM pretraining.
Unknown
The paper introduces corrective machine unlearning, a framework for updating models to remove or correct harmful data points without full retraining.
Unknown
The paper argues that implicit bias in learning dynamics controls neural scaling laws during training, linking growth to scaling behavior from perceptrons to deep networks.
Unknown
RocketKV is a training-free two-stage KV cache compression method that accelerates long-context LLM inference by combining coarse-grain permanent eviction with fine-grain dynamic sparsity.
Chen An, Fei Huang, Jun Zhang, et al.
Proposes a training-free method to extend the context window of large language models beyond their pretraining length without additional training.
Unknown
PostTrainBench benchmarks how well LLM agents can autonomously perform LLM post-training under bounded compute constraints (10 hours on one H100 GPU).
Unknown
The paper introduces a training approach to enhance fine-grained visual perception in vision-language models, reducing visual biases and improving detail recognition.
Unknown
The paper shows that vision foundation models can serve as effective tokenizers for latent diffusion models, but distillation-based tokenizer training weakens performance.
Unknown
A review of urban foundation models, covering pre-training methods and recent advances toward urban general intelligence.