This paper presents the benefits of visual pretraining for foundation model intelligence, outperforming text-only pretraining on multiple backbones and benchmarks.

This paper compares the performance of recurrent decoders and selective state-space models (Mamba) in intracortical brain-to-text systems, and investigates the impact of output targets (phonetic vs. c…

The paper introduces Brain-IT-VQA, a novel framework that significantly improves visual question answering from fMRI signals, and presents NSD-VQA, a new, highly controlled dataset for this task.

This paper proposes BrainAgent, an agentic LLM framework for knowledge-enhanced brain network analysis, which improves performance and produces more comprehensive, multi-level, and verifiable explanat…

The paper proposes MindDiffuser, a two-stage framework that significantly improves image reconstruction from brain activity by combining semantic guidance from text-to-image models with structural ref…

The paper introduces Eyettention II, a novel deep-learning model that can generate realistic, detailed scanpaths—including fixation location, within-word landing position, and duration—to address the…

Mind-Omni introduces a unified multi-task framework that models the interplay between brain, vision, and language signals using a discrete diffusion paradigm, achieving state-of-the-art performance ac…

This paper investigates the use of contrastive objectives for brain decoding using functional MRI (fMRI) activity and shows that linear contrastive decoders outperform other methods.

MindVoice is a neuro-to-speech framework that uses pretrained priors to disentangle and reconstruct intelligible speech from noisy, non-invasive neural signals, significantly outperforming existing me…

The authors introduce Structured PubMed, a comprehensive corpus of section-labeled biomedical abstracts compiled from the complete PubMed database.

PromptEmbedder introduces a dual-LLM framework that efficiently and transferably adapts text embeddings by decoupling task-specific knowledge from the backbone model, significantly reducing computatio…

This paper introduces SkMTEB, a comprehensive text embedding benchmark for Slovak, and develops efficient, locally-deployable Slovak embeddings.

This paper compares the preservation of linguistic content in different text encodings (tokens, bytes, pixels) using a shared bottleneck, revealing their distinct strengths in surface form preservatio…

This paper introduces KliniskVestBERT, a suite of BERT models specialized by pre-training on a large, diverse corpus of real-world Norwegian clinical texts, demonstrating superior performance for clin…

The paper proposes RTFree-F5, a method to make flow-matching TTS models like F5-TTS independent of reference transcripts, improving performance and naturalness for dysarthric speakers.

The paper proposes a graph-constrained approach to scale multi-hop training data by decoupling path discovery from path verbalization, significantly expanding the usable corpus size for LLMs.

This paper demonstrates that targeted adaptation of the small language model Phi Silica, using dataset curation and fine-tuning, significantly improves its performance in short-form text rewriting, na…

Pangram Labs introduces Pangram 4, an improved deep-learning-based AI-text classification model with increased accuracy, superior out-of-distribution generalization, and robustness to adversarial atta…