AI.info
Latest Research
Browse Latest Research on AI.info.
- Block Sparse Flash Attention
- Biases in the Blind Spot: Detecting What LLMs Fail to Mention
- SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Corpora
- Accelerating Vision Transformers with Adaptive Patch Sizes
- Softmax Reparameterization for Output-Head Quantization
- RaLiFlow: Scene Flow Estimation with 4D Radar and LiDAR Point Clouds
- HiPLEX: Hierarchical Policy Factorization for Full Duplex Speech Language Models
- Are they lovers or friends? Evaluating LLMs' Social Reasoning in English and Korean Dialogues
- Bootstrap Off-policy with World Model
- CTRL-ALT-DECEIT: Sabotage Evaluations for Automated AI R&D
- HistoTx: Early fusion of H&E images and spatial transcriptomics at varying spatial transcriptomics resolution with self-supervised learning
- MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues
- CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation
- WeVisDoc: From Coverage to Capability for Robust End-to-End Document Parsing
- DeepCompress: A Dual Reward Strategy for Dynamically Exploring and Compressing Reasoning Chains
- MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes
- It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them
- YANAMI: A Large-Scale Multimodal Dataset and Benchmark for Multiturn Stateful Interaction
- When a Conceptor Complement Becomes Gain: Operator Geometry and Behavioral Attribution in Language Models
- SimWAM: A Simple World Action Model for End-to-End Autonomous Driving
- Hierarchical Spatio-Temporal Transformer for Coherent Emergency Department Forecasting
- MieDB-100k: A Comprehensive Dataset for Medical Image Editing
- VDC-Agent: When Video Detailed Captioners Evolve Themselves via Agentic Self-Reflection
- Towards AI as Colleagues: Multi-Agent System Improves Structured Ideation Processes
- DRACO: Fine-Grained Credit Assignment with Dynamic Rubrics for Long-Horizon Agent Training
- FirstAidQA: A Synthetic Dataset for First Aid and Emergency Response in Low-Connectivity Settings
- $μ$pscaling small models: Principled warm starts and hyperparameter transfer
- On the Expressive Power of GNNs for Boolean Satisfiability
- LoPA: Scaling dLLM Inference via Lookahead Parallel Decoding
- 3DPR: Single Image 3D Portrait Relight using Generative Priors
- Bilevel Layer-Positioning LoRA for Real Image Dehazing
- EgoGeneralist: Unifying Egocentric Action-Video Tasks through Video-Native Modeling
- Stress Testing Factual Consistency Metrics for Long-Document Summarization
- An automatic counting algorithm for the quantification and uncertainty analysis of the number of microglial cells trainable in small and heterogeneous datasets
- PA-VAD: Diffusion-Based Pseudo-Only Video Anomaly Detection via Domain-Aligned Memory Updates
- Otter: Mitigating Background Distractions of Wide-Angle Few-Shot Action Recognition with Enhanced RWKV
- Feedback Control for Multi-Objective Graph Self-Supervision
- SVD-NO: Learning PDE Solution Operators with SVD Integral Kernels
- Finch: Benchmarking Finance & Accounting across Spreadsheet-Centric Enterprise Workflows
- Is Complex Agentic Memory Always Needed? Naive RAG Is Stronger Than We Thought