Book Review: ‘Minor Black Figures,’ by Brandon Taylor
nytimes.com·23h
🌐Network Protocols
Parallelizing Cellular Automata with WebGPU Compute Shaders
vectrx.substack.com·23h·
Discuss: Substack
🎨Graphics Programming
Constraint Satisfaction Approaches to Wordle: Novel Heuristics and Cross-Lexicon Validation
arxiv.org·5d
⚙️Compilers
Revisiting Mixout: An Overlooked Path to Robust Finetuning
arxiv.org·2d
⚙️Compilers
Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs
arxiv.org·4d
🧩programming
Bridging Reasoning to Learning: Unmasking Illusions using Complexity Out of Distribution Generalization
arxiv.org·2d
🧩programming
Detecting Distillation Data from Reasoning Models
arxiv.org·4d
🧩programming
Quantifying Data Contamination in Psychometric Evaluations of LLMs
arxiv.org·2d
🧩programming
Empowering LLMs with Pseudo-Untrimmed Videos for Audio-Visual Temporal Understanding
arxiv.org·2d
⚙️Compilers
Surgeons Are Indian Males and Speech Therapists Are White Females: Auditing Biases in Vision-Language Models for Healthcare Professionals
arxiv.org·2d
🎨Graphics Programming
Evolutionary Profiles for Protein Fitness Prediction
arxiv.org·2d
⚙️Compilers
Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning
arxiv.org·2d
⚙️Compilers
XRPO: Pushing the limits of GRPO with Targeted Exploration and Exploitation
arxiv.org·2d
⚙️Compilers
h1: Bootstrapping LLMs to Reason over Longer Horizons via Reinforcement Learning
arxiv.org·2d·
Discuss: Hacker News
🧩programming
FURINA: A Fully Customizable Role-Playing Benchmark via Scalable Multi-Agent Collaboration Pipeline
arxiv.org·2d
🧩programming
Visual Representations inside the Language Model
arxiv.org·4d
⚙️Compilers
EVALUESTEER: Measuring Reward Model Steerability Towards Values and Preference
arxiv.org·2d
⚙️Compilers
Off-Trajectory Reasoning: Can LLMs Collaborate on Reasoning Trajectory?
arxiv.org·2d
⚙️Compilers
On knot detection via picture recognition
arxiv.org·2d
🎨Graphics Programming
Fine-Tuning Large Language Models with QLoRA for Offensive Language Detection in Roman Urdu-English Code-Mixed Text
arxiv.org·4d
⚙️Compilers