Tag
Sergey Levine recommends the ACT or ALOHA paper for AI robotics beginners, highlighting how low-cost robot setups and simple transformer models can perform complex dexterous tasks.
This paper proposes a Cross-Platform Fairness Evaluation framework to audit transformer models for mental health NLP, revealing significant performance and calibration failures when models are applied across different social media platforms.
This paper introduces an AI-enhanced method for RF interference suppression using transformer models with a Finite Scalar Quantization tokenizer to boost performance and minimize latency, validated via audio metrics in digital signal scenarios.
The article explores optimizing AI language model performance on a home server built from e-waste GPUs, with explanations of transformer models and multi-GPU techniques.
ATHENA is a knowledge-guided agentic neural architecture search framework that automates Transformer-based electronic health record modeling by reusing architecture knowledge across hospitals to reduce manual tuning.
This paper investigates AI learning and conceptual transfer in the Game of Hidden Rules, focusing on reinforcement learning with Transformer-based A2C framework, rule difficulty analysis, transfer learning, and generalization.
This critical narrative review examines how large language models and transformer-based systems are reshaping hadith computational science, highlighting progress in data resources and segmentation tasks while identifying gaps in narrator verification and reproducibility, and proposing a research agenda for methodological strengthening.
A comparative study of BART, BERT, and RoBERTa for text summarization, examining their architectures and suitability for extractive and abstractive summarization tasks.
This paper introduces Strategic 16K, a leakage-controlled dataset for document sensitivity classification, and benchmarks classical and transformer-based models, with BERT achieving top performance while addressing label leakage issues.
This paper challenges the conflation of dynamic parameterization with dynamic inference, introducing Frozen-Controller Auditing to show that input-dependent coefficients do not imply computational savings. Experiments on Transformers reveal static layerwise profiles retain near-full performance despite no conditional execution.
This paper proposes modifications to encoder-decoder transformers for unconstrained molecular structure elucidation from IR spectra, using a Mixture-of-Experts decoder and contrastive alignment loss, achieving over 10 percentage points improvement in Top-K accuracy.
This paper investigates how instruction-tuned Transformer models (LLaMA and Mistral) encode the discourse relations of causation and antithesis using interpretability techniques on next-token prediction.
A curated list of resources for mastering LLM cache management, including explanations, tutorials, and research papers on KV cache, prefix caching, and related techniques.
Loopie is a new looped Transformer model that achieves gold-medal performance at the 2025 IMO and IPhO without external tools, using a novel post-training pipeline. It outperforms vanilla Transformers trained with the same compute budget.
DataStates-LLM introduces a scalable checkpointing architecture for transformer models using composable state providers, achieving up to 4x higher throughput and reducing training time by 2.2x compared to existing solutions.
This paper presents a systematic empirical study of fine-tuning pretrained Transformer models (Wav2Vec2.0, HuBERT, XLS-R) for Quranic Automatic Speech Recognition (ASR), achieving a WER of 0.08 on the EveryAyah subset and reducing training time from 140 to 40 hours, with Wav2Vec2-XLSR-53 providing the best representation.
Google researchers published a paper summarizing the evolution of TPU supercomputers from TPU v2 to Ironwood, detailing architectural stability, scale, resilience, power efficiency, and a 3600x performance increase over eight years.
This paper critically reviews hadith computational science in the era of large language models, highlighting progress, gaps in methodology, and proposing a research agenda to enhance evidence infrastructure for Islamic scholarship.
The author asks how to analyze the relative 'strength' of probes in neural networks, discussing challenges such as limited vocabulary size and model capacity, and using an example from Google Gemini to illustrate failure cases.
This paper introduces a dual diagnostic framework to trace the internal lifecycle of code reasoning in LLMs, revealing that models first 'brew' answers and then diverge into four resolution outcomes, with stable brewing across architectures but varying resolution success.