Knowledge catalogue
Hey all, I'm serving DSv4Flash 0731 on a cluster of 2x DGX Sparks but am running into constant issues with having almost no RAM (unified memory) left for the OS/cache and I'd love to hear the communit
arXiv:2506.18203v3 Announce Type: replace Abstract: Verifiers can improve language model capabilities by scoring and ranking responses from generated candidates. Currently, high-quality verifiers are
arXiv:2608.05734v1 Announce Type: new Abstract: Subliminal Learning (SL) is a surprising type of generalization displayed by modern language models. It allows the transfer of a bias or behavior from a
Alibaba’s Qwen team announced the new Qwen‑3.8‑Max model and thanked users for extensive testing, noting its ability to provide highly detailed world observations. A community member highlighted that
arXiv:2608.05423v1 Announce Type: cross Abstract: In generative training, a model produces an output and is penalised for its difference from an example. With one output per comparison, a model that p
arXiv:2608.04753v1 Announce Type: new Abstract: Attention layers are the backbone of today's most powerful and impactful models. Models with multi-million and billion parameters rely on contextual kno
arXiv:2602.06337v2 Announce Type: replace-cross Abstract: Causal inference is essential for decision-making but remains challenging for non-experts. While large language models (LLMs) show promise in
arXiv:2608.05149v1 Announce Type: new Abstract: Current instruction-based image retrieval systems are powerful but limited to single-turn interactions, failing to capture the iterative nature of compl
arXiv:2608.04956v1 Announce Type: new Abstract: Recent video models increasingly support generation, reference conditioning, and editing within a single model, yet typically expose them as separate op
arXiv:2509.24122v3 Announce Type: replace Abstract: At the heart of time-series forecasting (TSF) lies a fundamental challenge: how can models efficiently and effectively capture long-range temporal d
arXiv:2608.04095v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly used as personalized assistants in high-stakes domains such as financial advising, yet it remains unc
arXiv:2608.04374v1 Announce Type: cross Abstract: Large language models can produce fluent financial analysis, but fluency alone does not establish whether a report is suitable for institutional deliv
arXiv:2608.04505v1 Announce Type: new Abstract: This technical report presents K-EXAONE 2.0, an open-weight multilingual foundation model developed by LG AI Research as a step in our effort toward glo
arXiv:2608.04678v1 Announce Type: new Abstract: Papers 1-2 of the Kathleen series showed that a byte-level, attention-free architecture built from a wavetable encoder and multi-scale reverberant state
The quality of open-weight language models has dramatically improved in recent years. Sharing weights greatly facilitates model adoption by enabling their use across diverse hardware and software plat
arXiv:2608.05049v1 Announce Type: new Abstract: Instruction-based video editing (IVE) is an emerging field with broad applications, yet evaluating editing models remains challenging. Existing benchmar
arXiv:2608.04939v1 Announce Type: new Abstract: Social media videos often communicate meanings that go beyond their visible actions, captions, or speech. A mundane clip may become humorous, ironic, or
arXiv:2608.04783v1 Announce Type: cross Abstract: The integration of Large Language Models (LLMs) into software engineering has shifted the focus from function-level generation to repository-scale ass
arXiv:2602.11506v4 Announce Type: replace-cross Abstract: The transition toward localized intelligence through Small Language Models (SLMs) has intensified the need for rigorous performance characteri
arXiv:2603.18774v2 Announce Type: replace Abstract: Foundational feed-forward visual geometry models enable accurate and efficient camera pose estimation and scene reconstruction by learning strong sc
arXiv:2608.04415v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to detect unsafe content. A common approach is to combine judgments from a panel of models to correct
arXiv:2608.05138v1 Announce Type: cross Abstract: Modern Greek is absent from NVIDIA's Nemotron retrieval models and from major multilingual retrieval benchmarks, despite being important for retrieval
arXiv:2602.11745v2 Announce Type: replace Abstract: Graph models are fundamental to data analysis in domains rich with complex relationships. Unlike SQL, which benefits from a rel- atively unified sta
arXiv:2608.04726v1 Announce Type: new Abstract: Multimodal large language models increasingly reason over screenshots and documents where the task itself may be written in pixels. Yet benchmarks usual
arXiv:2608.03070v1 Announce Type: cross Abstract: Frontier AI model developers increasingly rely on layered safeguards to prevent catastrophic misuse, but little public evidence exists on how much pro
arXiv:2608.03464v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have made significant progress in chart understanding, generation, and editing, but their ability to annotate e
arXiv:2608.02678v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) systems are vulnerable to corpus poisoning: an attacker who inserts a crafted document into the retrieval corpus
arXiv:2608.03206v1 Announce Type: cross Abstract: Large language models (LLMs) power educational applications from tutoring to essay scoring, but each is a point solution to a single task, and only re
arXiv:2608.03259v1 Announce Type: cross Abstract: As time-series foundation models have emerged, the need for benchmarks that can evaluate their forecasting ability in meaningful ways has become incre
I built a local OCR pipeline a few days ago, and it turned into a surprisingly interesting experiment—taking accuracy from around 60% to 99%. I wrote a short blog about what worked, what failed, and t
arXiv:2608.02826v1 Announce Type: cross Abstract: Reinforcement learning is a subfield of machine learning that studies how an agent interacts with an environment in order to extract as large a reward
arXiv:2608.02915v1 Announce Type: cross Abstract: Domain-specific Instruction Set Architecture eXtensions (ISAX) are widely adopted in the RISC-V ecosystem to accelerate emerging workloads, but implem
As you all know the model is 2.69B parameters with a 128K context window and purpose-built for multi-step agent workflows. What you are seeing is the Q4_K_M GGUF running on my own inference engine bui
arXiv:2511.13300v1 Announce Type: cross Abstract: Generative models have shown remarkable performance in speech enhancement (SE), achieving superior perceptual quality over traditional discriminative
arXiv:2608.03341v1 Announce Type: cross Abstract: Large language models (LLMs) have substantially improved code generation, yet achieving strong functional correctness remains difficult, especially fo
arXiv:2608.02665v1 Announce Type: cross Abstract: A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form. We ask whether that reading is f
arXiv:2608.03291v1 Announce Type: cross Abstract: Chain-of-thought (CoT) reasoning improves large language model (LLM) performance while also providing an observable interface to the model's reasoning
arXiv:2608.03817v1 Announce Type: cross Abstract: Large vision--language models (LVLMs) demonstrate strong multimodal reasoning capabilities but remain prone to hallucination, where model predictions
arXiv:2608.02829v1 Announce Type: new Abstract: Model families train every size from scratch. Can a pretrained large model be converted into a smaller sibling? We characterize the 1.4B->410M conversio
arXiv:2608.01258v1 Announce Type: new Abstract: The realism of images generated by multimodal large language models (MLLMs), such as GPT Image2 and Nano Banana2, has improved rapidly in recent years.
arXiv:2608.00434v1 Announce Type: new Abstract: Multi-Token Prediction (MTP) has emerged as an effective paradigm that augments a shared Large Language Model backbone with auxiliary heads, training th
arXiv:2608.01865v1 Announce Type: new Abstract: Automatic speech recognition (ASR) performance degrades sharply on dysarthric speech, yet how disordered articulation reshapes a model's internal repres
arXiv:2608.02016v1 Announce Type: new Abstract: Sparse voxel grids preserve the spatial structure needed for detailed 3D reconstruction, but their memory still grows rapidly with resolution as active
arXiv:2608.00799v1 Announce Type: new Abstract: Computer-Aided Design (CAD) underpins modern engineering, yet converting existing shapes into editable models still demands substantial expert effort. M
arXiv:2608.01021v1 Announce Type: cross Abstract: In an age of rapid model turnover, how do we make hallucination evaluation more perennial? We explore whether human-written hallucination samples coul
arXiv:2602.01132v2 Announce Type: replace Abstract: Tasks such as solving arithmetic equations, evaluating truth tables, and completing syllogisms are handled well by large language models (LLMs) in t
arXiv:2603.12147v2 Announce Type: replace Abstract: Egocentric video provides a natural modality for studying human behavior, but conventional visual understanding captures mainly observable scenes, o
arXiv:2608.00909v1 Announce Type: new Abstract: Can large language models generate not just correct, but fast hardware? This paper investigates the question in financial FPGA design, where 5-10 nanose
arXiv:2607.09842v2 Announce Type: replace-cross Abstract: We investigate whether identity-specifying system prompts produce statistically distinguishable geometric fingerprints in the hidden-state tra
arXiv:2608.02068v1 Announce Type: new Abstract: Monocular depth foundation models, benefiting from large-scale synthetic training data, have demonstrated strong generalization. However, they often hal
arXiv:2603.15257v2 Announce Type: replace Abstract: Tactile sensing is a crucial capability for Vision-Language-Action (VLA) architectures, as it enables dexterous and safe manipulation in contact-ric
arXiv:2608.00852v1 Announce Type: new Abstract: Understanding and controlling complex dynamical systems often requires executing thousands of numerical simulations across vast parametric landscapes, w
Disclosure: I’m the author and maintainer of QuarkStar. I built QuarkStar, a small native inference engine inspired by Antirez’s DwarfStar. QuarkStar currently supports: Qwen3.6-35B-A3B, using the sam
arXiv:2608.01207v1 Announce Type: new Abstract: Test-time scaling lifts large language model reasoning by sampling many candidate solutions and selecting among them, yet the same recipe transfers poor
arXiv:2602.04789v4 Announce Type: replace Abstract: Advanced autoregressive (AR) video generation models have improved visual fidelity and interactivity, but the quadratic complexity of attention rema
arXiv:2608.02449v1 Announce Type: new Abstract: Deploying vision-language models (VLMs) for safety-critical spatial reasoning on resource-constrained autonomous driving platforms requires both compact
For robotaxis and other autonomous vehicles (AVs), the hardest problems aren’t the everyday scenarios. They’re the rare, complex situations that are difficult to anticipate and train for. Handling the
arXiv:2608.02372v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed in task-oriented dialogue systems that support multi-step decision-making in high-stakes domains
arXiv:2608.00687v1 Announce Type: new Abstract: LiDAR point clouds provide explicit, deterministic physical boundaries critical for collaborative safety-critical perception. However, wireless channels
Qwen 3.8‑Max, Alibaba.Qwen’s latest large‑language model, has been added to Hermes Agent and can currently be accessed at a 20 % discount. The update aims to streamline integration of the model for de