AI Research

Researchers introduced SearchEyes, a multimodal search agent that uses a typed knowledge graph to unify training data, search environments, and reward signals, achieving state-of-the-art performance on six benchmarks. Th…

Researchers introduced CanvasAgent, a multimodal agent that orchestrates multiple visual tools for complex image creation and editing, along with the CanvasCraft dataset containing 140K annotated trajectories. The agent …

Researchers developed a vision-language framework using Low-Rank Adaptation (LoRA) on a frozen BiomedCLIP backbone to detect severe adverse events in wound monitoring. The system integrates clinical notes and wound descr…

Researchers introduced Scene Graph Thinking (SaGe), a paradigm that enhances multimodal large language models by converting flat image-text data into structured scene graphs for fine-grained visual reasoning. The method,…

Researchers introduced Statistically Meaningful Geometry (SMG), a framework modeling over-parameterized learning systems as infinite-dimensional Orlicz fiber bundles, to differentiate genuine intelligence from statistica…

Researchers introduced Design-CP, two context-parallel inference strategies for RFdiffusion 3 that distribute quadratic activations across multi-GPU meshes, enabling all-atom design of large protein nanoparticles. The me…

A study on the 'Granularity Paradox' in time-series forecasting finds that finer temporal disaggregation improves in-sample fit but degrades out-of-sample accuracy due to recursive error compounding. Benchmarking 10 mode…

Researchers formalize the execution harness of LLM agents as a learnable control layer using offline reinforcement learning, training a lightweight controller with advantage-weighted regression to improve verification be…

Researchers introduced AdaStop, a cost-aware early stopping framework for deep neural network testing that determines when to stop labeling inputs by comparing the marginal fault discovery rate to a cost-benefit threshol…

Researchers introduced InvWeaver, a neuro-symbolic framework that synthesizes loop invariants for programs with multiple interacting loops by exposing inter-loop dependencies and propagating proof obligations. The system…

Researchers introduce Stochastic Token Steering (STS) and Stochastic Block Steering (SBS), which apply sparse autoencoder-based activation steering to only a fraction of tokens, reducing fluency degradation while maintai…

Researchers introduced Reflexion, an AI-powered prototype for structured emotional self-reflection that integrates emotion detection, layered prompting, and metaphorical storytelling. The system guides users from emotion…

Researchers developed a reduced NEXI protocol for quantifying human gray matter microstructure on the Connectome 2.0 scanner, using an Explainable AI framework to cut scan time from 27 to 14 minutes. Validated in seven p…

Researchers introduced a Tri-Branch Modular Fusion Neural Network that combines 3D geometry, SMILES embeddings, and physicochemical descriptors for molecular property prediction. The model achieved a validation MAE of 0.…

A new arXiv paper demonstrates that standard importance weighting fails to correct for selection bias in machine learning when selection depends on unobservables correlated with outcomes. The authors adapt Heckman's 1979…

Researchers derived smooth Monte-Carlo estimators for Minkowski functionals of neural fields, achieving 1-3% accuracy in 2D and 3D topology estimation at 250x speedup over persistent homology. However, in 3D neural-SDF f…

Researchers developed a power-calibrated statistical framework for logit-based LLM watermarking that establishes explicit relationships between hyperparameters, detection power, and distortion, transforming watermark des…

Researchers propose retrieval-augmented generation and constrained decoding to improve LLM-generated web API invocation code. RAG reduces hallucinations but can introduce unnecessary parameters, while constrained decodin…

Researchers show that incorrect L0 hyperparameter settings in Sparse Autoencoders (SAEs) cause feature mixing and degenerate solutions, preventing proper disentanglement of LLM internal features. They propose a proxy met…

Researchers proposed a method to estimate observation parameters and noise in inverse problems using a diffusion-based prior model within a Bayesian framework. The approach enables optimal estimation of both parameters a…