HarmoVid: Relightful Video Portrait Harmonization
Explorar
Noticias de IA
21271 elementos — filtrados, clasificados y sin duplicados
Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior …
Skill-Conditioned Gated Self-Distillation for LLM Reasoning
Reasoning that Travels: Dissecting How Chain-of-Thought Transfers Across Models
The Abstraction Gap in Vision-Language Causal Reasoning
LLM Zeroth-Order Fine-Tuning is an Inference Workload
NVIDIA Releases Polar, a Token-Faithful Rollout Framework for GRPO Training Across Codex,…
MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems
Stage-wise Distortion-Perception Traversal in Zero-shot Inverse Problems with Diffusion M…
Beyond Binary Moral Judgment: Modeling Ethical Pluralism in AI
Understanding Generalization and Forgetting in In-Context Continual Learning
History-aware adaptive reduced-order models via incremental singular value decomposition
Optimal ridge regularization revisited
GraphLit: Learning Text-Enriched Dynamic Character Network Representations for Literary S…
When Interpretability Is Unequally Distributed: Fairness in Hybrid Interpretable Models
Mining Multi-Modality Spatio-Temporal Cues for Video Important Person Identification
Position: Retire the "Positive Backdoor" Label -- Secret Alignment Requires Strict and Sy…
PLS in the Mirror of Self-Attention
Continual Model Routing in Evolving Model Hubs
Semantic Optimal Transport for Sparse Autoencoder Feature Matching and Circuit Compression
Refusal Before Decoding: Detecting and Exploiting Refusal Signals in Intermediate LLM Act…
GEM: Generative Supervision Helps Embodied Intelligence
DriveWAM: Video Generative Priors Enable Scalable World-Action Modeling for Autonomous Dr…
GUI-CIDER: Mid-training GUI Agents via Causal Internalization and Density-aware Exemplar …
Semi-Supervised Hypothesis Testing by Betting on Predictions
The Decision to Verify: How Warmth and User Characteristics Shape Reliance on Conversatio…
DiscoForcing: A Unified Framework for Real-Time Audio-Driven Character Control with Diffu…
Comonadic Morphophonology: A Compositional Framework for Context-Dependent Morphological …
The Cases LJP Never Sees: Prosecution Decision Prediction for More Complete Criminal Liab…
GONDOR to the Rescue: Satisficing Planning with Low Memory