RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation
Explorar
Noticias de IA
37834 elementos — filtrados, clasificados y sin duplicados
Seattle is close to approving a year-long ban on large data centers
Google's Backstops Underpin $35 Billion Anthropic Chip Deal
Layer-Isolated Evaluation: Gating the Deterministic Scaffold of a Production LLM Agent wi…
not much happened today
Can AI Reason Like an Urban Planner? Benchmarking Large Language Models Against Professio…
TSMC’s Monthly Sales Rise 30% Thanks to Sustained AI Chip Demand
Probabilistic Salary Prediction with Graph Attention Networks and a Mixture Density Netwo…
TreeSeeker: Tree-Structured Trial, Error, and Return in Deep Search
Sparse probes and murky physics: a case study of interpretability challenges in a foundat…
Tree-Structured Orthonormal Decomposition of the Aitchison Simplex
Improving Cross-Format Robustness in Language Models with Multi-Format Training
Attacks on Machine-Text Detectors Retain Stylistic Fingerprints
Einride Joins US-Listed Driverless Trucking Rivals on Nasdaq
A Survey of Robotic Navigation and Manipulation with Physics Simulators in the Era of Emb…
CleanPatrick: A Benchmark for Image Data Cleaning
Recoverable but Not Stationary:Local Linear Structures in Weights and Activations
Transformer Based Model for Spatiotemporal Feature Learning in EEG Emotion Recognition
Detecting Knowledge Gaps from Conversational AI Interactions Using Curriculum Prerequisit…
GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerat…
EstRTL: Functional Estimation Guided RTL Code Generation
Spatial-Omni: Spatial Audio Understanding Integration in Multimodal LLMs via FOA Encoding
CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs
A History-Aware Visually Grounded Critic for Computer Use Agents
CLP: Collocation-Length Prediction for Zero-Loss Adaptive Multi-Token Inference
Democratising Camera Trap AI: An Open-Source Model for Detecting UK Mammals
Minimum Distortion Quantization with Specified Output Distribution
A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in…
Beyond Uniform Token-Level Trust Region in LLM Reinforcement Learning
Understanding and mitigating the risks of OpenClaw for non-technical users: A practical g…