Knowledge catalogue
arXiv:2609.20068v1 Announce Type: new Abstract: This paper builds a theoretical bridge between the economic notion of marginal utility and two machine-learning constructs, matrix factorization and the
arXiv:2609.20604v1 Announce Type: new Abstract: This paper presents a real-time semantic world modeling framework specialized for precision agriculture using autonomous robots. The framework combines
arXiv:2609.18005v1 Announce Type: new Abstract: Large language model optimization is an active research area, spanning quantization of model weights, early-exit methods for skipping layers, and specul
arXiv:2609.18935v1 Announce Type: new Abstract: A game character should not have to reread its entire life before every conversation. For locally deployed language-model characters, however, revising
arXiv:2505.03590v2 Announce Type: replace-cross Abstract: Magnetic resonance spectroscopy (MRS) is a non-invasive technique to measure the metabolic composition of tissues, offering valuable insights
arXiv:2609.17815v1 Announce Type: cross Abstract: The Koopman operator has been widely used for time-series prediction in dynamical systems. However, prior work that learns latent ``Koopman spaces'' u
arXiv:2609.17204v1 Announce Type: cross Abstract: Wi-Fi signal data can be used to compromise the privacy of individuals. While many existing approaches rely on Channel State Information (CSI), collec
arXiv:2609.17384v1 Announce Type: new Abstract: Robot teams that learn a common environment model exchange belief summaries and plan by the expected information gain of their actions. Under conjugate
arXiv:2609.15527v1 Announce Type: new Abstract: The energy efficiency of analog computing makes it one of the most promising candidates for deploying resource-intensive machine learning workloads on c
arXiv:2606.22906v2 Announce Type: replace-cross Abstract: Large language models have shown strong performance on software engineering (SE) tasks, yet understanding large industrial repositories remain
arXiv:2609.13682v1 Announce Type: cross Abstract: We introduce LayerRoute, a parameter-efficient method for adaptive transformer layer-skipping that combines per-layer hard-gated routing (trained via
arXiv:2604.05793v2 Announce Type: replace-cross Abstract: In LLM agent pipelines, prompt privacy risk propagates beyond a single model call: raw user content enters retrieval queries, memory writes, t
arXiv:2609.12825v1 Announce Type: new Abstract: State Space Models (SSMs), particularly VMamba, have emerged as efficient alternatives for modeling long-range dependencies in medical image analysis. H
arXiv:2609.07022v1 Announce Type: cross Abstract: Reasoning over privacy-constrained records requires combining structured attributes with evidence from free-text narratives. We introduce CIPHER (Cros
arXiv:2609.05774v1 Announce Type: new Abstract: Recent multi-agent LLM systems increasingly rely on graph-structured communication to coordinate specialized agents. We revisit multi-agent orchestratio
arXiv:2609.09682v1 Announce Type: new Abstract: Many learning systems return a single point estimate even when admissible outputs form disconnected or non-convex sets. Under squared loss, an ambiguous
arXiv:2609.10355v1 Announce Type: cross Abstract: Video understanding has rapidly evolved toward video large language models (VideoLLMs): systems that couple video representations with pretrained larg
arXiv:2510.23258v2 Announce Type: replace-cross Abstract: Autonomous robotic navigation in real-world environments requires exploration to acquire environmental information as well as goal-directed na
arXiv:2609.08234v1 Announce Type: cross Abstract: Suppose we are given an ordered sequence of independent data whose distribution changes K times at unknown locations, for some unknown K geq 0. In thi
arXiv:2609.07403v1 Announce Type: new Abstract: Privacy-sensitive surveillance systems could benefit from large vision-language models (VLMs), but such models typically require centralized access to r
The fix: run the server detached/headless instead of keeping it attached to a console window. RTX 5090, ~27B NVFP4 model via ninfer: Terminal focused: 130–200 tok/s Terminal unfocused: 50–60 tok/s Cli
Benchmark two 30B Mixture-of-Experts models, Qwen3-Coder-30B and NVIDIA Nemotron-3-Nano-30B, across G5, G6, G6e, and G7 GPU instances on Amazon SageMaker AI. Compare throughput, latency, and cost-per-
arXiv:2609.04895v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models activate only a small subset of experts per token, but the full expert set often exceeds GPU memory, causing repeated we
arXiv:2606.19667v2 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) improves factual grounding, but it also lengthens prompts and raises prefill cost. Prefix caching in serving en
arXiv:2609.04653v1 Announce Type: new Abstract: Multilingual large language models often struggle to reason in low- to mid-resource languages. Prior work has shown that translation can improve multili
arXiv:2609.05275v1 Announce Type: new Abstract: Layer dropout (a.k.a. stochastic depth) has been shown to enable faster training, higher accuracy, and robustness to zero-shot layer pruning in both lan
arXiv:2609.05060v1 Announce Type: cross Abstract: As quantum hardware scales to larger devices, the classical software layers that interface with it must evolve in step. Postprocessing routines develo
arXiv:2609.03844v1 Announce Type: cross Abstract: We introduce Stateless Bernoulli Watermarking (SBW), a new statistical watermark for Large Language Models that determines green list membership throu
arXiv:2504.02694v3 Announce Type: replace-cross Abstract: We study counterfactual regression, which maps features to outcomes under hypothetical scenarios that differ from those observed in the data.
arXiv:2609.04005v1 Announce Type: new Abstract: We present an alternative characterization of the occupancy measure of reinforcement learning, obtained by embedding the planning criterion into the dyn
I'm one of the developers. We said in August it would go open source in September and it did last night. MIT or Apache-2.0, pick one. The repo you see is our internal repo, kernels included, so from n
arXiv:2609.03554v1 Announce Type: cross Abstract: Generative retrieval has demonstrated significant success by unifying representation learning and search into a single sequence-to-sequence generation
arXiv:2609.01723v1 Announce Type: cross Abstract: Text-to-Speech (TTS) foundation models are increasingly fine-tuned on private datasets to synthesize highly personalized voices, introducing severe pr
arXiv:2609.02302v1 Announce Type: new Abstract: A core obstacle to alignment evaluation is evaluation awareness: capable models can tell when they are being tested rather than deployed, weakening the
arXiv:2609.01335v1 Announce Type: cross Abstract: Many production systems can assess a configuration only by using it on live requests and observing noisy feedback. Modern agentic systems are a promin
arXiv:2609.00363v1 Announce Type: new Abstract: Conformance suites for quantized GEMM kernels ask whether two implementations agree within a tolerance. We measure what such a suite can detect. Injecti
arXiv:2609.00450v1 Announce Type: cross Abstract: Block Quantization (BQ) is a promising approach for efficient deployment of large language models (LLMs), enabling low-precision computation with cont
arXiv:2609.00873v1 Announce Type: new Abstract: Diffusion language models (DLMs) have recently emerged as an alternative modeling paradigm to autoregressive LMs, offering advantages such as parallel g
arXiv:2505.13299v2 Announce Type: replace-cross Abstract: This paper considers the estimation of quantiles via a smoothed version of the stochastic gradient descent (SGD) algorithm. By smoothing the s
arXiv:2609.01277v1 Announce Type: cross Abstract: Although pretrained joint audio-visual diffusion models offer rich control over what to generate, they provide no explicit control over when an uttera
arXiv:2608.28691v1 Announce Type: cross Abstract: Wearable VLM pipelines promise continuous multimodal assistance from egocentric visual capture: a user asks a task-driven question about the surroundi
arXiv:2608.28886v1 Announce Type: cross Abstract: What does a generation loop gain from learning on its own verified successes? In cycles of generate, verify, select and LoRA-consolidate on online bin
arXiv:2608.28726v1 Announce Type: new Abstract: The remarkable performance of multimodal large language models (MLLMs) comes at the cost of substantial computational overhead, posing significant chall
I only started using some vLLM forks recently in a 4 x 48GB 4090 system. DS4F - ~5000pp/180tg (DSpark) Qwen3.8 Flash next - ~7500pp/135tg (MTP) This is amazing, like having the API in my house. But it
arXiv:2608.30505v1 Announce Type: cross Abstract: Large language models (LLMs) are built from structured high-dimensional objects such as token representations, weights, adaptation updates, caches, an
arXiv:2305.19366v3 Announce Type: replace Abstract: Generative Flow Networks (GFlowNets), a class of generative models over discrete and structured sample spaces, have been previously applied to the p
arXiv:2608.27080v1 Announce Type: cross Abstract: Many scientific and engineering applications require estimating unknown parameters from experimentally observable data -- an inverse problem that is i
arXiv:2608.26674v1 Announce Type: cross Abstract: As large language models are increasingly deployed to simulate diverse human characters, ensuring persona fidelity, defined as the extent to which an
arXiv:2608.25181v1 Announce Type: new Abstract: Collective dynamics arise in a wide range of physical, biological, and engineering applications. Examples include cell migration, swarm robotics, social
arXiv:2608.23986v1 Announce Type: cross Abstract: Large language model providers are compute constrained, and their universal response to congestion is to degrade service: route queries to smaller mod
arXiv:2608.22024v1 Announce Type: cross Abstract: Estimating heterogeneous single and interaction treatment effects from observational data under multiple simultaneous treatments is crucial for decisi
arXiv:2608.22652v1 Announce Type: cross Abstract: LLMs are increasingly used for code generation, yet they frequently hallucinate non-existent software packages, creating exploitable entry points into
arXiv:2608.21362v1 Announce Type: new Abstract: Transformer-based large language models (LLMs) incur high prefill latency because key-value (KV) tensors must be recomputed for each request. Existing p
arXiv:2608.22643v1 Announce Type: cross Abstract: Deploying large language models on edge devices is increasingly limited by a widening gap between model size and available memory. Existing approaches
We have been working on some performance optimisations for Qwen3.8 and other models. The main new feature that we introduced is adaptive speculation for Llama.cpp What is it? MTP and DFlash work well
arXiv:2506.13583v2 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) agents often exhibit learning behaviors that are not intuitively interpretable by human observers, which can resul
arXiv:2606.13629v2 Announce Type: replace-cross Abstract: There is a proliferation of work arguing for the use of synthetic data in scientific research. For example, social scientists are arguing for
arXiv:2608.20026v1 Announce Type: new Abstract: Streetscape quality has become a central concern in contemporary urban planning, particularly within the framework of the pedestrian-friendly 15-minute
Benchmarks: Benchmark U1 Preview Full Qwen-Image-Bench 47.14 55.20 (PE) 60.18 (PE) ImgEdit 3.9 4.37 4.59 GEdit-Bench-EN 7.47 8.14 8.26 Instead of just scaling up, they train task-specialized expert mo
arXiv:2608.19553v1 Announce Type: new Abstract: Vision--language models can identify the correct referent while returning an imprecise bounding box. We study whether a frozen direct-answer model can u