The paper investigates whether natural language text possesses an intrinsic curvature, proposing a new metric called Texture that captures word-level discrete curvature. Texture is defined as a signed two-axis curvature of the word-in-context belief field, measuring how context from one side contracts or expands the semantic effect of context from the other side. The authors provide empirical and theoretical evidence of non-flat semantic inference, define Texture formally, and demonstrate its practical utility in improving long-context inference and retrieval-augmented generation.
arXiv:2609.24564v1 Announce Type: new Abstract: CLIP, a foundational vision-language model, has emerged as a powerful tool for open-vocabulary semantic segmentation. While freezing CLIP's text encode...
arXiv:2609.01147v1 Announce Type: cross Abstract: Text-rich visual inputs require models that can read, retrieve, and compress language directly in pixel space, yet existing pixel-text encoders strug...
GraSP‑VL demonstrates that the length of frozen vision‑language embeddings can serve as a controllable semantic interface. By learning a shared near‑orthogonal prefix transform, the method creates a Semantic Matryoshka where short prefixes encode coarse semantics and longer prefixes reveal finer language‑grounded distinctions, all while preserving the original embedding geometry. Experiments on COCO/Flickr30K and SugarCrepe‑clean show strong performance with negligible drift in the full embedding space.
arXiv:2607. 07047v1 Announce Type: cross Abstract: Understanding the geometric structure of pre-trained language model embeddings matters for interpretability and safety.
arXiv:2609.37230v1 Announce Type: cross Abstract: Visual distinctions are often finer than those reflected in linguistic conceptualization. Vision-language models exhibit a similar asymmetry: a disti...
arXiv:2607. 20092v1 Announce Type: cross Abstract: Contextual entrainment is the tendency of a model to let auxiliary context in its input pull its output, independently of whether that context is relevant, true, or even meaningful.
arXiv:2605. 08245v4 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) increasingly power high-stakes applications, from medical imaging to autonomous systems, yet they routinely hallucinate, confidently describing content not present in the input.
arXiv:2605. 13178v2 Announce Type: replace-cross Abstract: In large vision-language models, visual tokens typically constitute the majority of input tokens, leading to substantial computational overhead.
The paper introduces TIC‑Bench, a new benchmark for evaluating multimodal large language models on deeply interleaved text‑image contexts. It covers logical, temporal, and spatial association tasks, totaling 2,280 questions across eight specific types. The authors benchmarked ten state‑of‑the‑art MLLMs, finding a significant performance gap versus human experts and highlighting persistent challenges in integrating evidence across interleaved visual and textual inputs.
arXiv:2608. 07886v1 Announce Type: cross Abstract: Vision-language grounding connects language to visual content, yet most existing formulations reduce grounding to a unidirectional localization problem: given a prespecified text phrase or category name, identify the corresponding image region.
The article presents a technical manual for an open toolkit designed to measure how transformer language models individuate word meanings across different contexts. It introduces the concept of a "bridge form"—a single word that appears unchanged in multiple domains but with distinct senses—and outlines a full pipeline from specifying these forms to extracting layer-wise representations, computing silhouette-based separation metrics, and visualizing results. The manual details each design choice and its intended methodological safeguards, emphasizing that it serves as a methodological reference rather than reporting empirical findings.
ASCIIBench is a new benchmark that evaluates large language models on generating and classifying ASCII-text images, using a dataset of 5,315 labeled ASCII images. The authors also release a fine‑tuned CLIP model adapted to capture ASCII structure for evaluation. Their analysis shows that cosine similarity on CLIP embeddings fails to separate most categories, indicating a representation bottleneck rather than generational variance.