AI.info
Latest Research — Page 9
Browse Latest Research on AI.info.
- Beyond Teacher Likelihood: Group-Calibrated On-Policy Distillation for Long-Context Reasoning
- IntroConformal: Conformal Factuality Guarantees for Large Vision-Language Models via Introspective Signals
- From Native Memes to Global Moderation: Cross-Cultural Evaluation of Vision-Language Models for Hateful Meme Detection
- RcAE: Recursive Reconstruction Framework for Unsupervised Industrial Anomaly Detection
- Contextual and Seasonal LSTMs for Time Series Anomaly Detection
- Retrieval Augmented Generation Compositionality
- APPLE: Attribute-Preserving Pseudo-Labeling for Diffusion-Based Face Swapping
- Hierarchical Sparse Plus Low Rank Compression of LLM
- A receding-horizon multi-contact motion planner for legged robots in challenging environments
- DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences
- Tractable Multinomial Logit Contextual Bandits with Non-Linear Utilities
- Fairness Without Labels: Pseudo-Balancing for Bias Mitigation in Face Gender Classification
- On-the-Fly OVD Adaptation with FLAME: Few-shot Localization via Active Marginal-Samples Exploration
- Latent Introspection: Models Can Detect Prior Concept Injections
- Breaking the Vicious Cycle: Coherent 3D Gaussian Splatting from Sparse and Motion-Blurred Views
- Scaling in Games: Continued Pre-Training For Embodied Agents In Diverse Virtual Worlds
- Efficient Training for Human Video Generation with Entropy-Guided Prioritized Progressive Learning
- EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI
- Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning
- Many Right Answers, One Training Example: Removing Responder Noise from GRPO for LLM Recommenders
- Ponimator: Unfolding Interactive Pose for Versatile Human-human Interaction Animation
- Inverse Autoregressive Flows for Zero Degree Calorimeter fast simulation
- ChipBench: A Next-Step Benchmark for Evaluating LLM Performance in AI-Aided Chip Design
- Autonomous Concept Drift Threshold Determination
- PARSE: LLM Driven Schema Optimization for Reliable Entity Extraction
- Mull-Tokens: Modality-Agnostic Latent Thinking
- Art2Mus: Artwork-to-Music Generation via Visual Conditioning and Large-Scale Cross-Modal Alignment
- IndEgo: A Dataset of Industrial Scenarios and Collaborative Work for Egocentric Assistants
- ARROW: An Adaptive Rollout and Routing Method for Global Weather Forecasting
- OmniEdit-Bench: A Comprehensive Benchmark for Instruction-based Video Editing
- AtlasNLP: A Country-Aware Atlas of Dataset Representation in NLP
- Text Has Curvature
- Annotrace: A Workbench for Step-Level Annotation of Agent Traces
- Learning Native Continuation for Action Chunking Flow Policies
- PACT: Can Enterprise AI Assistants Be Trusted Under Pressure?
- LayerEdit: Disentangled Multi-Object Editing via Conflict-Aware Multi-Layer Learning
- Systematicity between Forms and Meanings across Languages Supports Efficient Communication
- UTF-8 Plumbing: Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8
- Ontology Learning with LLMs: A Benchmark Study on Axiom Identification
- Miner:Mining Intrinsic Mastery for Data-Efficient RL in Large Reasoning Models