About
Research & Development
Stress-Testing LLMs: New Benchmark Reveals Fragility in Mathematical Reasoning
TLDR: A new research paper introduces the Generalization–and–Perturbation (GAP) framework and PutnamGAP dataset to evaluate LLM mathematical reasoning robustness. By creating mathematically equivalent but...
Research & Development
Unpacking Cultural Biases Within AI Language Models
TLDR: This research introduces CultureScope, a novel method to investigate how large language models (LLMs) internally process and represent cultural knowledge. It reveals that...
Research & Development
Advancing Causal Machine Learning Through Better Synthetic Experiments
TLDR: A research paper argues that rigorous synthetic experiments are crucial for the broader adoption of Causal Machine Learning (Causal ML). It identifies key...
Research & Development
Improving Arabic Voice Recognition Through Weak Supervision and Fine-tuning
TLDR: A new Arabic Automatic Speech Recognition (ASR) system combines weakly supervised pretraining on 15,000 hours of diverse Arabic speech with continual supervised fine-tuning...
Research & Development
EGGCodec: Advancing F0 Extraction Through Robust EGG Signal Reconstruction
TLDR: EGGCodec is a new neural framework designed for precise electroglottography (EGG) signal reconstruction and fundamental frequency (F0) extraction. It improves upon existing models...
Research & Development
Teaching LLMs to Be Concise: A New Approach to Efficient Reasoning
TLDR: A new curriculum learning strategy for large language models (LLMs) called "Train Long, Think Short" uses Group Relative Policy Optimization (GRPO) to improve...