AI.info

Latest Research — Page 125

Browse Latest Research on AI.info.

• CoPRS: Learning Positional Prior from Chain-of-Thought for Reasoning Segmentation
• Culturally-Aware Conversations: A Framework & Benchmark for LLMs
• Evidence-Grounded Verified Agentic Reasoning: A Path Toward Eliminating LLM Hallucination in Empirical Inference via Tool-Attested Kernel Proofs
• Bridging the Gap Between Molecule and Textual Descriptions via Substructure-aware Alignment
• ClaimDB: A Fact Verification Benchmark over Large Structured Data
• General and Efficient Steering of Diffusion Models
• Memory-V2V: Memory-Augmented Video-to-Video Diffusion for Consistent Multi-Turn Editing
• GloTok: Global Perspective Tokenizer for Image Reconstruction and Generation
• AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
• FlareX: A Physics-Informed Dataset for Lens Flare Removal via 2D Synthesis and 3D Rendering
• OneThinker: All-in-one Reasoning Model for Image and Video
• Test-Time Adaptation by Causal Trimming
• Convergence Theorems for Entropy-Regularized and Distributional Reinforcement Learning
• 1+1>2: A Synergistic Sparse and Low-Rank Compression Method for Large Language Models
• Stable Velocity: A Variance Perspective on Flow Matching
• AutoFly: Vision-Language-Action Model for UAV Autonomous Navigation in the Wild
• Autonomous Mathematical Discovery in an Open-World Multi-Agent Environment
• Native Reasoning Models: Training Language Models to Reason on Unverifiable Data
• Instruction-Driven 3D Facial Expression Generation and Transition
• On-Demand Lecture Watching System Using Various Actions of Student Characters to Maintain Concentration
• Efficient Restarts in Non-Stationary Model-Free Reinforcement Learning
• eTracer: Towards Traceable Text Generation via Claim-Level Grounding
• ToxicTextCLIP: Text-Based Poisoning and Backdoor Attacks on CLIP Pre-training
• CostNav: A Navigation Benchmark for Real-World Economic-Cost Evaluation of Physical AI Agents
• SynWeather: Weather Observation Data Synthesis across Multiple Regions and Variables via a General Diffusion Transformer
• SynAD: Enhancing Real-World End-to-End Autonomous Driving Models through Synthetic Data Integration
• CountSteer: Steering Attention for Object Counting in Diffusion Models
• Key and Value Weights Are Probably All You Need: On the Necessity of the Query, Key, Value weight Triplet in Self-Attention Transformers
• Mapping Human Anti-collusion Mechanisms to Multi-agent AI Systems
• MASA: LLM-Driven Multi-Agent Systems for Autoformalization
• Enhancing Foundation Models in Transaction Understanding with LLM-based Sentence Embeddings
• Hide&Seek: Learning to Explain in an End-to-End Differentiable Network
• Quantifying Epistemic Uncertainty in Diffusion Models
• Unified Interactive Multimodal Moment Retrieval via Cascaded Embedding-Reranking and Temporal-Aware Score Fusion
• Your Autoregressive Model Already Reveals the Causal Graph
• MovieRecapsQA: A Multimodal Open-Ended Video Question-Answering Benchmark
• AdaSearch: Balancing Parametric Knowledge and Search in Large Language Models via Reinforcement Learning
• TRIM: Scalable 3D Gaussian Diffusion Inference with Temporal and Spatial Trimming
• What, Where, and How: Disentangling the Roles of Task, Language, and Model in Code Model Representations
• MASCOT: Multi-Agent Socio-Collaborative Companion Systems