VIHD: Visual Intervention-based Hallucination Detection for Medical Visual Question Answe…
Explorar
Noticias de IA
27413 elementos — filtrados, clasificados y sin duplicados
Diffuse to Detect: Bi-Level Sample Rebalancing with Pseudo-Label Diffusion for Point-Supe…
MTR-Suite: A Framework for Evaluating and Synthesizing Conversational Retrieval Benchmarks
Self-Training Doesn't Flatten Language -- It Restructures It: Surface Markers Amplify Whi…
The Readout Shortcut: Positional Number Copying Dominates Arithmetic CoT Readout in Small…
An OpenAI model has disproved a central conjecture in discrete geometry
CASCADE Conformal Prediction: Uncertainty-Adaptive Prediction Intervals for Two-Stage Cli…
Spectral Souping: A Unified Framework for Online Preference Alignment
Decomposing MXFP4 quantization error for LLM reinforcement learning: reducible bias, reco…
Less Data, Faster Training: repeating smaller datasets speeds up learning via sampling bi…
Stage-adaptive Token Selection for Efficient Omni-modal LLMs
CAMERA: Adapting to Semantic Camouflage in Unsupervised Text-Attributed Graph Fraud Detec…
optimize_anything: A Universal API for Optimizing any Text Parameter
Closed-form predictive coding via hierarchical Gaussian filters
When the Majority Votes Wrong, the Intervention Timing for Test-Time Reinforcement Learni…
LambdaPO: A Lambda Style Policy Optimization for Reasoning Language Models
You Don't Need Attention: Gated Convolutional Modeling for Watch-Based Fall Detection
Matérn Noise for Triangulation-Agnostic Flow Matching on Meshes
Cross-Paradigm Knowledge Distillation: A Comprehensive Study of Bidirectional Transfer Be…
FPED: A Functional-Network Prior-Guided Mixture-of-Experts Framework for Interpretable Br…
CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering
Fast-tracking genetic leads to reverse cellular aging
PIXLRelight: Controllable Relighting via Intrinsic Conditioning
General Preference Reinforcement Learning
EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL
SPIKE: An Adaptive Dual Controller Framework for Cost-Efficient Long-Horizon Game Agents
Import AI 457: AI stuxnet; cursed Muon optimizer; and positive alignment
Recent Developments in LLM Architectures: KV Sharing, mHC, and Compressed Attention
Finding the molecular switches behind new infectious diseases
Opening new paths in aging research