Retrieval pipelines, vector search, chunking and reranking: how models are grounded in a corpus instead of their weights.
arXiv:2607. 03682v1 Announce Type: cross Abstract: Convection-dominated convection-diffusion problems often develop thin layers, where the solution has sharp transition profiles and its derivatives are highly localized.
arXiv:2508. 04928v5 Announce Type: replace-cross Abstract: We propose a method to extend foundational monocular depth estimators (FMDEs), trained on perspective images, to fisheye images.
arXiv:2607. 03447v1 Announce Type: cross Abstract: Knowledge graphs (KGs) that underpin Graph-based Retrieval-Augmented Generation (Graph-RAG) are increasingly built automatically by LLM-driven extraction rather than curated by experts.
arXiv:2607. 05185v1 Announce Type: new Abstract: Compositional generalization, the ability to understand and produce novel combinations of known components, remains a fundamental challenge for modern artificial intelligence.
Recent advancements in Multimodal Large Language Models (MLLMs) have evolved from static perception to interleaved visual-language reasoning, often referred to as ``thinking with images''. A basic operation in this reasoning process is to zoom in on regions of interest (often represented with bounding boxes) to acquire finer visual details.
Large-scale Vision-Language Models like CLIP have demonstrated impressive open-set localization capabilities at the image level. However, adapting this capability to pixel-level dense prediction poses challenges due to global feature biases.
Audio intelligence involves understanding, reasoning about, and generating both audio and speech. In this work, we introduce Nemotron-Labs-Audex-30B-A3B (Audex), a unified audio-text LLM built on Nemotron-Cascade-2-30B-A3B, a strong text-only MoE LLM.
Enterprise Document Intelligence [Vol. 1 #8C] - Structured output is the start of validation, not the end: check the evidence, accept not-found, loop the feedback The post Validating the RAG Answer Before the User Sees It: Spans, Quotes, and the Feedback Loop appeared first on Towards Data Science .
Joint-Embedding Predictive Architectures (JEPAs) underpin a growing family of latent world models for control from raw pixels, but every existing JEPA world model commits at training time to a single inference paradigm: either trajectory optimisation in a learned dynamics model, or direct behaviour cloning. A single checkpoint that serves both would defer this choice to inference, when deployment constraints (rollout cost, observation accessibility) determine which path wins.
Zero-shot composed image retrieval (ZS-CIR) aims to retrieve a target image from a multimodal query consisting of a reference image and an edit text describing the desired modification. Recent ZS-CIR studies have relied on projection-based methods that map a reference image into pseudo-word tokens in the text embedding space.
Retrieval-Augmented Generation (RAG) streamlines long-document understanding by leveraging retrieval mechanisms to restrict input images to a highly curated subset. However, existing multimodal RAG pipelines primarily face two critical challenges: first, standard semantic similarity retrievers frequently fetch topically overlapping yet answer-void distractor pages that mislead downstream generation; second, rigid single-pass pipelines heavily depend on initial retrieval success, where any omission of core evidence inevitably causes cascading errors.
Accurate prediction of patient-specific therapeutic response from pre-treatment transcriptomes is hindered by the scarcity of matched clinical response labels and post-treatment molecular profiles. Preclinical transfer-learning models can simulate drug-induced expression changes but are often hard to interpret and unstable, whereas knowledge-graph methods provide mechanistic context yet remain static and fail to capture drug-induced transcriptomic perturbation dynamics.
AI reasoning has become a central focus in contemporary artificial intelligence, largely driven by the success of large language models. However, mathematical research, which is characterized by non-linear derivation paths, rigorous logical requirements, and protracted exploration cycles, poses severe challenges for existing reasoning systems.
Enterprise Document Intelligence [Vol. 1 #8B] - A fixed BASE, the rules each question needs, one registry: the dispatcher that turns a parsed question into a typed LLM call The post Assemble Each RAG Generation Prompt from a Base Prompt Plus the Rules Each Question Needs appeared first on Towards Data Science .
Enterprise Document Intelligence [Vol. 1 #8A] - The schema is the contract: every field is a question the pipeline asks the model, and every answer is checkable The post Stop Returning Text from RAG: The Typed Answer Contract That Prevents Hallucination appeared first on Towards Data Science .
Most "LLM wikis" use agents, embeddings, and repeated model calls to organize local notes. I built a deterministic alternative: a pure Python compiler that turns messy markdown into a linked, linted wiki using only the standard library.
Enterprise Document Intelligence [Vol. 1 #7ter] - Six positions on the retrieval brick that contradict the cosine-first reflex of mainstream RAG The post The Untaught Lessons of RAG Retrieval: Cosine Is Not the Foundation appeared first on Towards Data Science .
arXiv:2607. 01455v1 Announce Type: cross Abstract: Language models learn continuous programs over discrete symbols, with the embedding table and LM-head acting as the read/write interface between them.
arXiv:2607. 02212v1 Announce Type: cross Abstract: Aqueous solubility is a key property in early-stage drug discovery, but most predictive models merge physicochemical descriptors and molecular graph information into a single representation, obscuring whether a prediction is driven by global chemistry, molecular structure, or both.
arXiv:2512. 22227v3 Announce Type: replace-cross Abstract: We investigate whether graded states of mind form spectrum-like structure in transformer representation spaces.