AI.info

Latest Research — Page 56

Browse Latest Research on AI.info.

• Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability
• Can Fine-Tuning Erase Your Edits? On the Fragile Coexistence of Knowledge Editing and Adaptation
• GYROval: A Robust Benchmark for Cultural Value Orientation in Large Language Models
• SpatialReward: Bridging the Perception Gap in Online RL for Image Editing via Explicit Spatial Reasoning
• Neural Additive Experts: Context-Gated Experts for Controllable Model Additivity
• Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection
• TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems
• BanglaWild: An In-the-Wild Bengali Scene Text Recognition Benchmark for OCR and Vision-Language Models
• Stratified Knowledge-Density Super-Network for Scalable Vision Transformers
• Partial Action Replacement: Tackling Distribution Shift in Offline MARL
• ClaimPT: A Portuguese Dataset of Annotated Claims in News Articles
• Hard Labels In! Rethinking the Role of Hard Labels in Mitigating Local Semantic Drift
• Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated?
• Sample Complexity of Distributionally Robust Off-Dynamics Reinforcement Learning with Online Interaction
• MAT-Agent: Adaptive Multi-Agent Training Optimization
• EasyTune: Efficient Step-Aware Fine-Tuning for Diffusion-Based Motion Generation
• Towards Valid B-Rep Generation: Training-Free Wireframe Anomaly Detection and Repair
• Metacognitive Sensitivity for Test-Time Dynamic Model Selection
• Unsupervised Learning of Efficient Exploration: Pre-training Adaptive Policies via Self-Imposed Goals
• WildSEEK: Evaluating Language Models for Information-Seeking
• Gradient-Weight Alignment as a Train-Time Proxy for Generalization in Classification Tasks
• Old Habits Die Hard: How Conversational History Geometrically Traps LLMs
• Mitigating Error Accumulation in Co-Speech Motion Generation via Global Rotation Diffusion and Multi-Level Constraints
• Diff-ICMH: Harmonizing Machine and Human Vision in Image Compression with Generative Prior
• Zoom-IQA: Image Quality Assessment with Reliable Region-Aware Reasoning
• Invisible Triggers, Visible Threats! Road-Style Adversarial Creation Attack for Visual 3D Detection in Autonomous Driving
• Hash Collisions in Molecular Fingerprints: Effects on Property Prediction and Bayesian Optimization
• FlexAvatar: Learning Complete 3D Head Avatars with Partial Supervision
• Finite-Probe Total-Variation Certificates for Finite-Basis Drifting Models
• OWL: Unsupervised 3D Object Detection by Occupancy Guided Warm-up and Large Model Priors Reasoning
• Boosting Reasoning in Large Multimodal Models via Activation Replay
• Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance
• G-Net: A Provably Easy Construction of High-Accuracy Random Binary Neural Networks
• Hyperball May Not Be a Free Lunch
• Zero-Shot Mission-Level Evaluation for Aerial MLLM Agents
• Modeling Hierarchical Thinking in Large Reasoning Models
• Surface-based Molecular Design with Multi-modal Flow Matching
• When Eyes and Ears Disagree: Can MLLMs Discern Audio-Visual Confusion?
• Downscaling Intelligence: Exploring Perception and Reasoning Bottlenecks in Small Multimodal Models
• Soft Shadow Diffusion (SSD): Physics-inspired Learning for 3D Computational Periscopy