I got Kimi-k3 running.....
Results: prompt eval: 40 tokens / 97.5s → 0.41 tok/s eval: 400 tokens / 1769.9s → 0.23 tok/s total: 440 tokens / 1867s (31 min) Prompt: 'Write a C++ function that reverses a linked list in place. Expl
Knowledge catalogue
Results: prompt eval: 40 tokens / 97.5s → 0.41 tok/s eval: 400 tokens / 1769.9s → 0.23 tok/s total: 440 tokens / 1867s (31 min) Prompt: 'Write a C++ function that reverses a linked list in place. Expl
If I only went off X posts, I'd think Ramp was an AI lab We’re open-sourcing PorTAL, our framework for shared task representations and cross model LoRA adaptation. It now spans from hybrid attention m
arXiv:2607.23322v1 Announce Type: new Abstract: Instruction tuning has become the standard method for adapting large language models to follow human intent, yet existing instruction datasets are domin
arXiv:2607.23503v1 Announce Type: cross Abstract: Internet of Things (IoT) applications generate vast amounts of Correlated Time Series (CTS) data that often contain missing values and require imputat
arXiv:2607.23808v1 Announce Type: cross Abstract: In this work, we introduce Indic DiarBench, a speaker diarization and ASR benchmark dataset spanning all 22 scheduled languages of India. This corpus
arXiv:2601.05680v2 Announce Type: replace-cross Abstract: While Transformer-based autoregressive models excel in data generation, their token discretization strategy inherently limits their precision
arXiv:2607.24273v1 Announce Type: new Abstract: Large Language Models (LLMs) have shown strong potential in financial reasoning, but existing benchmarks often evaluate domain knowledge, numerical reas
Interesting. Grok 4.6 releases around August 7. This will be the 1.5T model with significantly improved SFT & RL. Grok 4.7 will be the 2.1T model released a few weeks later. This will be better than 4
arXiv:2607.23078v1 Announce Type: new Abstract: Longitudinal medical imaging captures temporal evolution of lesions, yet extracting the underlying dynamical parameters governing this evolution remains
It’s been five years since ChatGPT was released, AGI is not coming, not even the most modest AI predictions will ever become a reality, and there needs to be some of severe criminal sanction applied t
arXiv:2607.22783v1 Announce Type: cross Abstract: Recent advances in conventional and learning-based image coding have increased the demand for benchmark datasets that support fine-grained assessment
arXiv:2607.24405v1 Announce Type: new Abstract: In this work, we propose K-SurvMeans, a novel extension of K-Means for clustering survival data. The method explicitly uses the survival outcome in the
K3 is a language model that entered the top five most‑liked models on Hugging Face merely 24 hours after its release. The achievement surprised many, placing it ahead of prominent models such as Llama
arXiv:2607.23116v1 Announce Type: cross Abstract: KAYROS is an open-source solver for duration-minimization time-dependent vehicle routing problems, with or without time windows (TDVRPTW, TDVRP). In t
arXiv:2607.24368v1 Announce Type: new Abstract: Long-term memory systems store what a user says in an external store and retrieve it when a related query arrives. This interface rests on an assumption
arXiv:2607.23393v1 Announce Type: new Abstract: Existing exact methods for 4-connected grid pathfinding reduce online search, but often either retain fine-grained search states or require substantial
arXiv:2607.24653v1 Announce Type: new Abstract: We introduce Kimi K3, a 2.8T parameter Mixture-of-Experts model with 104 billion activated parameters, native vision capabilities, and a 1-million-token
arXiv:2607.23704v1 Announce Type: cross Abstract: The deployment of embodied agents in self-driving laboratories could accelerate scientific discovery, yet their reliability is constrained by the irre
arXiv:2607.22841v1 Announce Type: cross Abstract: We present DS@GT's submission to FinMMEval 2026 Task 1, a multilingual financial exam question answering benchmark spanning English, Spanish, Greek, C
arXiv:2607.23545v1 Announce Type: new Abstract: Instruction hierarchy (IH) requires models to prioritize instructions by source, ensuring that higher-priority instructions override lower-priority ones
arXiv:2607.22924v1 Announce Type: new Abstract: In the real world, fashion is about layering: adding a jacket over a shirt, or a sequence of adding and removing layers, rather than just a single-layer
arXiv:2607.22690v1 Announce Type: new Abstract: Long-term memory lets LLM agents reuse past interactions, but raw dialogue histories are verbose and information-sparse. Retrieving broadly improves evi
arXiv:2607.22777v1 Announce Type: cross Abstract: Protein language models learn transferable sequence representations. However, because they primarily model contextual dependencies along amino-acid se
arXiv:2607.23515v1 Announce Type: new Abstract: Long-horizon manipulation tasks pose significant challenges for reinforcement learning due to sparse reward signals and long horizons. Automatic curricu
arXiv:2607.23488v1 Announce Type: cross Abstract: Text-to-image diffusion models expose many inference-time sampling parameters, including prompts, negative prompts, classifier-free guidance scales, a
arXiv:2607.22748v1 Announce Type: cross Abstract: Modern neural networks primarily adapt through parameter modification within predefined computational structures. While recent methods introduce modul
arXiv:2507.05515v4 Announce Type: replace Abstract: Vision-language models (VLMs) are facing the challenges of understanding and following multimodal assembly instructions, particularly when fine-grai
LFM2.5-Encoder is a family of multilingual bidirectional encoders built on the LFM2 architecture, available in two sizes: LFM2.5-Encoder-230M — a lightweight encoder for tight latency and memory budge
arXiv:2411.00918v5 Announce Type: replace-cross Abstract: Mixture of experts (MoE) architectures have become a cornerstone for scaling up and are a key component in most large language models such as
arXiv:2607.24551v1 Announce Type: new Abstract: Maintenance regulations are complex legal texts that are difficult to exploit when addressing a specific case and challenging to integrate into operatio
arXiv:2607.24573v1 Announce Type: new Abstract: Large language models (LLMs) increasingly support decisions about uncertain future events, yet evaluating their ability to forecast real-world outcomes
arXiv:2607.24155v1 Announce Type: new Abstract: Existing research on affect in speech has shown how acoustic surface characteristics and content-related linguistic aspects of speech both relate to per
arXiv:2509.03059v2 Announce Type: replace-cross Abstract: Recent advances in Large Language Models (LLMs) have shown that their reasoning capabilities can be significantly improved through Reinforceme
arXiv:2607.23083v1 Announce Type: new Abstract: Gender-inclusive language generation seeks to transform biased text into inclusive alternatives while preserving semantic meaning and contextual coheren
https://github.com/woct0rdho/transformers5-qwen3.5-recipe An update on my progress with low-VRAM LoRA training over GGUF base model: Now we can train DeepSeek-V4-Flash (284B-A13B) in 90 GiB VRAM, with
arXiv:2607.24135v1 Announce Type: new Abstract: Single-image self-supervised denoising replaces unavailable clean targets with surrogate targets constructed from noisy observations. Its effectiveness
arXiv:2607.24518v1 Announce Type: new Abstract: Symmetric non-negative matrix factorization (SymNMF) recovers latent group structure from a dependence matrix, but its dense, quadratic-memory objective
arXiv:2607.22707v1 Announce Type: new Abstract: Single-image reflection removal aims to recover a clean transmission layer from one image captured through glass. We study an explicit decomposition pip
arXiv:2607.24101v1 Announce Type: cross Abstract: Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models. Existing evaluations
arXiv:2601.17039v2 Announce Type: replace-cross Abstract: Mangroves are critical for climate-change mitigation, requiring reliable monitoring for effective conservation. While deep learning has emerge
arXiv:2607.24224v1 Announce Type: new Abstract: Multi-modality data from different sensors provides rich complementary information for 3D perception, becoming an essential component in reliable autono
arXiv:2607.24424v1 Announce Type: new Abstract: Vision-language models commonly project all tokens produced by a pretrained vision encoder into a large language model. However, final-layer features ca
arXiv:2607.23785v1 Announce Type: cross Abstract: The Simple Temporal Problem (STP) is a core framework for quantitative temporal constraints. As STP data can be inconsistent, we study MAXSTP: compute
arXiv:2507.17636v2 Announce Type: replace Abstract: Negative campaigning is a defining feature of electoral competition, yet comparative research on its drivers has remained limited by the high cost a
arXiv:2607.15166v2 Announce Type: replace Abstract: Most medical AI benchmarks measure whether a model knows the correct answer. MedFailBench asks a different question: which safety boundary failed? W
arXiv:2607.22566v1 Announce Type: new Abstract: MedLoCoMo is a Medical Long-Context Memory benchmark for patient-specific clinical reasoning over multi-admission medical dialogue. Existing medical QA
arXiv:2607.24314v1 Announce Type: new Abstract: Predicting the absorption, distribution, metabolism, excretion and toxicity (ADMET) properties of small molecules remains a major challenge in drug disc
arXiv:2607.22696v1 Announce Type: cross Abstract: High-resolution video diffusion models built on Diffusion Transformers (DiTs) deliver strong fidelity but quickly exhaust the memory budget of a singl
arXiv:2607.23811v1 Announce Type: cross Abstract: Siri Expressive Voices synthesize rich, configurable speech in real time and entirely on device, powered by AFM 3 Core Advanced, Apple's most powerful
arXiv:2607.22890v1 Announce Type: cross Abstract: Domain Randomization (DR) is a standard technique for closing the Sim-to-Real gap, yet traditional DR pipelines rely on classical computer graphics re
Mage-VL is a codec-native, proactive-streaming multimodal foundation model for image and video understanding, whose visual encoder is trained entirely from scratch at a compact 4B scale. It targets a
Hello guys. I found some problems in current frontier models. And want to share. # math_code_hallucination > Record of a failure caused by combining mathematics and code in a single prompt. --- ## Cas
arXiv:2607.22556v1 Announce Type: new Abstract: Continual learning (CL) is essential for small language models (SLMs) to adapt to evolving real-world needs in resource-constrained deployments. However
arXiv:2607.23047v1 Announce Type: cross Abstract: Mixed-precision quantization improves the accuracy of post-training quantization by allocating higher bitwidths to sensitive layers, but existing meth
arXiv:2607.24407v1 Announce Type: new Abstract: Multimodal Large Language Models have made great progress in grounding tasks, yet existing methods still struggle to unify precise localization and comp
arXiv:2607.24665v1 Announce Type: new Abstract: Modern large language models scale successfully by pairing capacity growth with efficiency, keeping per-token and deployment costs under control as capa
arXiv:2607.22951v1 Announce Type: cross Abstract: Reliability assessment of large language models (LLMs) seeks to estimate the probability that a model produces correct responses under a specified ope
arXiv:2510.17604v2 Announce Type: replace Abstract: GNSS suffers from multipath errors in urban canyons, making reliable bicycle localization difficult. Hand-crafted inertial alternatives, such as cyc
arXiv:2607.24030v1 Announce Type: new Abstract: Massively multilingual automatic speech recognition (ASR) models covering hundreds of languages must maintain robust performance across diverse linguist
arXiv:2607.23870v1 Announce Type: cross Abstract: Smart-city airspace is transforming Uncrewed Aerial Vehicles (UAVs) from passive sensing platforms into cyber-physical decision makers that must follo