Knowledge catalogue
arXiv:2608.07921v1 Announce Type: cross Abstract: We apply Marchenko-Pastur (MP) random matrix theory to pre-trained attention weights in order to separate each projection matrix into a random-like bu
arXiv:2608.07712v1 Announce Type: cross Abstract: A predictive model receives a self-supervised signal whenever the consequence of an action is observed. Using that signal after deployment is difficul
arXiv:2601.09974v2 Announce Type: replace Abstract: Personalizing Large Language Models typically relies on static retrieval or one-time adaptation, assuming user preferences remain invariant over tim
arXiv:2608.08282v1 Announce Type: new Abstract: Tool-using language-model agents face constraints whose meaning changes with observations and prior actions. We study exact sampling from the model dist
arXiv:2608.08125v1 Announce Type: new Abstract: Recent advances in generative models have achieved remarkable performance in text- and image-conditioned editing. However, preserving the content of a g
Stealing Reasoning Traces from Proprietary LLM APIs A vanity domain name (stolen-thoughts.com) for a neat paper: Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that
arXiv:2608.09126v1 Announce Type: new Abstract: In this project, we delved into the pervasive challenge of bias detection within the text content. More specifically, our focus lies on the identificati
arXiv:2510.14357v2 Announce Type: replace-cross Abstract: Agricultural robots are emerging as powerful assistants across a wide range of agricultural tasks, nevertheless, they are still heavily relyin
arXiv:2505.11480v4 Announce Type: replace-cross Abstract: Superoptimization is the task of transforming a program into a faster one, and ideally the very fastest possible one, while preserving its inp
arXiv:2608.08253v1 Announce Type: new Abstract: AI agents are becoming shared infrastructure, yet durable memory is commonly assembled from separate retrieval, governance, and operational components.
arXiv:2608.07862v1 Announce Type: new Abstract: Existing safety evaluation datasets for large language models (LLMs) predominantly focus on English and Western contexts, often overlooking the linguist
arXiv:2608.08070v1 Announce Type: new Abstract: Reliable surgical planning requires models that move beyond recognizing the current surgical step or imitating expert demonstrations, and instead antici
arXiv:2608.07641v1 Announce Type: new Abstract: The rapid advancement of large language models has transformed survey writing from a months-long manual effort into an automated process. As generation
arXiv:2608.09802v1 Announce Type: new Abstract: As AI coding agents take on increasingly complex, long-horizon software engineering tasks, existing benchmarks are rapidly saturating and their evaluati
arXiv:2608.09497v1 Announce Type: new Abstract: Operational crop mapping requires models that generalise across years, resolve fine-grained crop taxonomies, and distinguish cropland from surrounding l
arXiv:2608.09162v1 Announce Type: cross Abstract: Tabular data presents unique challenges for deep learning due to its heterogeneous nature, where numeric features exhibit diverse distributions, scale
On August 11 2026, Gary Marcus commented that if the symbolic components were removed from Claude’s Riemann implementation it would fail to work, highlighting a concrete win for hybrid neurosymbolic A
arXiv:2608.09258v1 Announce Type: new Abstract: Multi-robot systems can accomplish tasks that are difficult for a single robot by organizing into task-specific formations. Different from existing stud
arXiv:2608.08528v1 Announce Type: new Abstract: Enterprise AI coding assistants incur substantial inference spend, and naive token-cost minimization often fails to reduce end-to-end cost once retries,
arXiv:2608.09538v1 Announce Type: cross Abstract: We introduce TCS-Bench, a benchmark for evaluating Large Language Models (LLMs) on research-level Theoretical Computer Science (TCS) proof generation.
arXiv:2608.07899v1 Announce Type: new Abstract: Agent systems increasingly expose execution traces, yet telemetry that reveals a failure may still be inadequate for identifying where that failure orig
arXiv:2608.07851v1 Announce Type: cross Abstract: Residual connections rely on a static residual pathway, and are essential for training deep neural networks. Hyper-connections (HC) increase the expre
arXiv:2608.07567v1 Announce Type: cross Abstract: Functional near-infrared spectroscopy (fNIRS) is a promising modality for autism spectrum disorder (ASD) classification, yet existing approaches assum
arXiv:2608.09393v1 Announce Type: cross Abstract: We identify and quantify temporal misgrounding: the systematic retrieval and citation of the currently in-force version of a legal article when the ap
arXiv:2602.00884v2 Announce Type: replace Abstract: Neural operators have shown promise in learning solution maps of partial differential equations (PDEs), but they often struggle to generalize when t
arXiv:2608.08290v1 Announce Type: new Abstract: Open-vocabulary semantic segmentation (OVSS) repurposes a pretrained CLIP encoder for dense prediction without additional labeled supervision. Existing
I'm guessing that many people have been waiting for this comparison. For clarity, both models are running at full FP16 KV-cache. Due to VRAM limitations, Muse Glimmer is running full 262,144 context,
arXiv:2608.08809v1 Announce Type: new Abstract: A single model scale challenges the flexibility of a production retrieval system: some settings need it faster, others need a smaller index, and the rig
arXiv:2608.07617v1 Announce Type: new Abstract: Scientific and technical writing depends on markup sources that must compile: LaTeX, Typst, and Markdown pipelines fail on missing delimiters, mismatche
⚙️That is the framework Mistral is building toward: one in which enterprises, governments, and startups can use the best AI available, shape it around their own knowledge, and retain the value it crea
arXiv:2608.08026v1 Announce Type: new Abstract: We investigate how social authority (SA) signals interact with severity-based prioritization in large language models, operationalizing each axis as a m
arXiv:2505.18347v3 Announce Type: replace-cross Abstract: Continual reinforcement learning (RL) concerns agents that are expected to learn continually, rather than converge to a policy that is then fi
The ChatGPT desktop app is now available in preview for desktop variants of these Linux distributions: • Ubuntu 24.04 LTS and 26.04 LTS • Debian 13 • Fedora 43 and 44 Install with .deb or .rpm package
arXiv:2511.02687v2 Announce Type: replace Abstract: The trajectory of AI development suggests that we will increasingly rely on agent-based systems powered by language models, composed of independentl
arXiv:2608.08826v1 Announce Type: new Abstract: Adaptive procedures must work without nuisance information an oracle may use, such as a gradient scale or smoothness index, and robust procedures may ha
arXiv:2608.08650v1 Announce Type: new Abstract: Mixture-of-Experts models increase parameter capacity while keeping the computation activated by each token bounded, but their architectural evolution c
arXiv:2608.07517v1 Announce Type: cross Abstract: Can a multimodal LLM predict which version of a web page will win a real A/B test from screenshots alone? We report the most complete answer we are aw
arXiv:2608.07528v1 Announce Type: new Abstract: Linear probes detect corrupted context in language models with near-perfect accuracy, yet this does not translate into reliable failure prediction. The
arXiv:2608.09574v1 Announce Type: new Abstract: LLMs are rapidly embedding themselves into daily life: drafting our emails, managing our schedules, and making decisions on our behalf. As they move fro
arXiv:2608.08654v1 Announce Type: new Abstract: How much an AI coding agent costs to run can depend more on the agent scaffolding that drives it than on the interface through which it reaches its tool
Imagine if your everyday laptop could run an AI model smart enough to take care of 90% of your work—totally private, lightning fast, and completely free of monthly fees. That is the exact tipping poin
💡The world needs an open-source platform, and that’s exactly what we’re building to give our customers more choice and the flexibility to choose the right model for the right task. As part of this, we
arXiv:2608.08881v1 Announce Type: new Abstract: The current work developed seven Retrieval-Augmented Generation (RAG) models based on leading deception theories and compared how deception judgments we
arXiv:2608.08113v1 Announce Type: new Abstract: Chain-of-Thought (CoT) prompting has become the dominant paradigm for eliciting reasoning in Large Language Models (LLMs), yet it creates substantial co
arXiv:2608.08168v1 Announce Type: new Abstract: While Large Language Models (LLMs) employing Chain-of-Thought (CoT) exhibit superior reasoning capabilities, the neural mechanisms distinguishing this e
This is part of working with the EU AI Act, other labs are adding similar watermarking. It’s hard to identify AI-generated text, and this gives people better tools to do that. We’ll also be a shipping
arXiv:2608.08910v1 Announce Type: new Abstract: PTQTP decomposes LLM weight matrices into two ternary (trit) planes with two free per-group scales. Tying the scales to a fixed ratio of three collapses
arXiv:2608.08512v1 Announce Type: new Abstract: Evolving documents, such as laws, tax codes, and software documentation, are amended, replaced, and sometimes reverted over time, so a question has diff
arXiv:2604.00715v2 Announce Type: replace-cross Abstract: Retrieval-augmented generation (RAG) improves language model (LM) performance by providing relevant context at test time for knowledge-intensi
I’ve been using ChatGPT for about two months now after giving up on Claude and Gemini and I became a true zealot. I’m now using Pro, and despite having the personality, set to default, never had any p
arXiv:2608.08727v1 Announce Type: cross Abstract: To address this gap, we introduce TomaMMU, a large-scale Tomato leaf disease MultiModal Understanding dataset, alongside TomaBench, a benchmark for ev
arXiv:2608.07917v1 Announce Type: new Abstract: Chinese historical documents preserve valuable cultural heritage, but many collections remain accessible only as scanned page images, preventing full-te
arXiv:2608.08951v1 Announce Type: new Abstract: Palm-vein recognition is a fine-grained biometric task in which both local vascular texture and the global layout of the vessel tree carry discriminativ
arXiv:2608.08844v1 Announce Type: new Abstract: Surgical instrument segmentation is a fundamental task for computer-assisted interventions, yet most existing methods rely on pixel-level annotations or
arXiv:2608.08795v1 Announce Type: cross Abstract: Tool-using large language model (LLM) agents are vulnerable to indirect prompt injection (IPI), in which malicious instructions embedded in external o
arXiv:2608.08885v1 Announce Type: cross Abstract: Reggaeton is one of the most widely consumed music genres in the world, and its lyrics are commonly regarded as highly sexualized. This claim rests mo
arXiv:2608.09861v1 Announce Type: new Abstract: Audio-visual interaction is the standard for patient-physician consultations, enabling natural communication and effective assessment of illness through
arXiv:2608.07700v1 Announce Type: new Abstract: Translating a natural-language question into a SPARQL query that can be executed against a large knowledge graph requires resolving lexical ambiguity, g
Not even sure if I'm allowed to post this, what with the 'completely/primarily LLM generated copy' rule (the post itself is fine, but the repo/model I'm sharing definitely is, whoops) and the whole li
arXiv:2608.09153v1 Announce Type: new Abstract: Production AI agents fail when their context sources -- system prompts, knowledge bases, tool descriptions, and procedural skills -- contain errors or g