AI.info

Latest Research — Page 80

Browse Latest Research on AI.info.

• CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization
• An Approach for Detection of Entities in Dynamic Media Contents
• Sigmoid Attention as a Better Substrate for Learned KV Cache Eviction
• A Primer on SO(3) Action Representations in Deep Reinforcement Learning
• Reliability Challenges in Diffusion Vision-Language Models
• LLEMA: Evolutionary Search with LLMs for Multi-Objective Materials Discovery
• Semantic Router: On the Feasibility of Hijacking MLLMs via a Single Adversarial Perturbation
• DeltaDorsal: Enhancing Hand Pose Estimation with Dorsal Features in Egocentric Views
• Now You See Me, Now You Don't: A Unified Framework for Expression Consistent Anonymization in Talking Head Videos
• AdaFuse: Adaptive Ensemble Decoding with Test-Time Scaling for LLMs
• Neural operator learning for collision-aware trajectory planning of spacecraft swarms
• The World Is Bigger! A Computationally-Embedded Perspective on the Big World Hypothesis
• RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation
• PASs-MoE: Mitigating Misaligned Co-drift among Router and Experts via Pathway Activation Subspaces for Continual Learning
• MoCo: A One-Stop Shop for Model Collaboration Research
• The Shape of Reasoning: Topological Analysis of Reasoning Traces in Large Language Models
• Steering Large Reasoning Models towards Concise Reasoning via Flow Matching
• Teaching According to Students' Aptitude: Personalized Mathematics Tutoring via Persona-, Memory-, and Forgetting-Aware LLMs
• Calibration Across Layers: Understanding Calibration Evolution in LLMs
• Accident Anticipation via Temporal Occurrence Prediction
• Pluralistic Behavior Suite: Stress-Testing Multi-Turn Adherence to Custom Behavioral Policies
• Measuring what Matters: Construct Validity in Large Language Model Benchmarks
• Adaptively Coordinating with Novel Partners via Learned Latent Strategies
• CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining
• FlowMimic: Mask-free Visual Editing and Generation with Pixel-pair Warped Flow Field for Online Video Editing Data Generation and Modality Mimicry
• Threshold Differential Attention for Sink-Free, Ultra-Sparse, and Non-Dispersive Language Modeling
• Cross-Modal Representational Knowledge Distillation for Enhanced Spike-Informed LFP Modeling
• Beyond Pairwise: Empowering LLM Alignment With Ranked Choice Modeling
• AIM: Anchor Identity Features, Then Match for Multimodal Large Language Model Unlearning
• Preliminary Analysis and Simulation of a Compact Variable Stiffness Wrist
• Uncertainty Quantification in LLM Agents: Foundations, Emerging Challenges, and Opportunities
• Efficient-SAM2: Accelerating SAM2 with Object-Aware Visual Encoding and Memory Retrieval
• GauDP: Reinventing Multi-Agent Collaboration through Gaussian-Image Synergy in Diffusion Policies
• Understanding Temporal Logic Consistency in Video-Language Models through Cross-Modal Attention Discriminability
• ImageSentinel: Protecting Visual Datasets from Unauthorized Retrieval-Augmented Image Generation
• OmniNWM: Omniscient Driving Navigation World Models
• Rethinking Reasoning with MDLMs: Early Exits, Post-hoc Reasoning, and Beyond
• Aerial Layouting: Design and Control of a Compliant and Actuated End-Effector for Precise In-flight Marking on Ceilings
• Dual-branch Spatial-Temporal Self-supervised Representation for Enhanced Road Network Learning
• Generalizing GNNs with Tokenized Mixture of Experts