Knowledge catalogue
arXiv:2609.35168v1 Announce Type: new Abstract: Saved checkpoints record states along a training trajectory, but generally do not determine the updates at states that would be visited under a differen
arXiv:2609.34497v1 Announce Type: new Abstract: Flow policies can model rich action distributions, but their iterative sampling limits decision speed. Adjoint matching uses the critic's action gradien
arXiv:2609.35581v1 Announce Type: cross Abstract: We introduce QC-Stark, a benchmark for evaluating large language models (LLMs) on 11 quantum computing (QC) tasks, spanning circuit construction, debu
arXiv:2609.34842v1 Announce Type: new Abstract: Existing multimodal time series foundation models (TSFMs) typically model heterogeneous modalities through largely shared mechanisms, overlooking the di
arXiv:2609.33088v1 Announce Type: new Abstract: Traditional change detection (CD) identifies changes between bi-temporal remote sensing images, while semantic change detection (SCD) assigns predefined
arXiv:2609.32455v1 Announce Type: new Abstract: Camera pose estimation from multi-view images remains a challenge in computer vision. Traditional methods often address this problem using Structure-fro
arXiv:2609.35094v1 Announce Type: new Abstract: Uncrewed aerial manipulators (UAMs) integrate robotic arms with aerial platforms for three-dimensional physical interaction. However, enlarging the work
arXiv:2609.34718v1 Announce Type: new Abstract: Reinforcement learning (RL) changes not only what language models say, but also how much they say, often increasing response length at the cost of token
arXiv:2609.34680v1 Announce Type: new Abstract: Four-bit post-training quantization can reduce the memory demands of large language models, but preserving accuracy under strict MXFP4 W4A4 requires coo
arXiv:2609.33638v1 Announce Type: new Abstract: We introduce RareTrap, a framework for estimating the probability of severe behaviors in black box large language models (LLMs). A key challenge for pro
arXiv:2509.23762v4 Announce Type: replace-cross Abstract: Spiking Neural Networks (SNNs) have recently received increasing attention in both computational neuroscience and artificial intelligence owin
arXiv:2601.04301v3 Announce Type: replace Abstract: As frontier AI systems are pretrained on web-scale data, test set contamination has become a critical concern for accurately assessing their capabil
arXiv:2609.34061v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models integrate pretrained Vision-Language Models (VLMs) with action heads for robot control. Common action heads have d
arXiv:2609.32659v1 Announce Type: new Abstract: Quantization-Aware Pre-Training (QAPT) can increase the inference efficiency of DNNs, but a problematic behaviour known as rounding boundary weight osci
arXiv:2609.33923v1 Announce Type: cross Abstract: A single random Gaussian probe gives an unbiased estimate of the squared Frobenius norm of a layer's quantization error. The estimator is well-behaved
arXiv:2609.35349v1 Announce Type: new Abstract: The evaluation cost of transformers with softmax attention scales quadratically with sequence length. Kernel attention addresses this by replacing softm
arXiv:2609.33097v1 Announce Type: new Abstract: Recent large-scale Vision-and-Language Navigation (VLN) models deliver strong accuracy but remain costly to deploy due to their large parameter counts a
arXiv:2609.31668v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) process multimodal inputs by converting text, images, and videos into token sequences that are subsequently pro
arXiv:2609.34273v1 Announce Type: cross Abstract: The projection of queries and keys are central to the attention mechanism in Transformer architectures. While they are mathematically symmetric, they
We evaluate several models on 100 tasks from the [internal Binary Exploitation benchmark] (selected at random), and find that GLM-5.3 develops full control flow hijacks in 4% of the trials; Claude Myt
arXiv:2609.33351v1 Announce Type: cross Abstract: Fidelity-based quantum retrieval ranks candidates by the fidelity between query and archive states. Applying a shared input-independent unitary after
I'm running Qwen 3.8 27B Q4 XS with ~30 t/s decode (no MTP) at 100k context with q8/q5 KV cache on a 16GB AMD GPU. I wanted to share my setup since many believe Qwen 3.8 27B to be infeasible on 16GB V
I started mapping the building blocks shared across all the models in audio.cpp. The result ended up being more interesting than I expected. Qwen has become by far the most common language backbone in
Qwen3.8-Flash is 40% off through the rest of the month, making now a perfect time to run your evals. Designed for high-volume applications like coding and coworking assistants, @Alibaba_Qwen's model o
arXiv:2609.33848v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly undertake extreme-long (xlong) horizon tasks, where a single execution can span hours, hundreds of model-
arXiv:2510.13854v2 Announce Type: replace-cross Abstract: We introduce Rule-to-Tag (R2T), a framework that turns linguistic rules into the training signal for neural sequence taggers in low-resource l
arXiv:2609.32836v1 Announce Type: cross Abstract: Radiomap blind prediction aims to infer radiomaps from observable representations of the propagation environment and base station configuration withou
arXiv:2609.33340v1 Announce Type: new Abstract: Contextual specialization can improve forecasting accuracy, but a correction selected on one historical interval may become unreliable under temporal di
arXiv:2609.34179v1 Announce Type: cross Abstract: Retrieval-augmented generation systems are extensively instrumented with metrics, benchmarks, traces, and automated judges, but these tools do not dec
arXiv:2609.33796v1 Announce Type: cross Abstract: Translating natural-language access-control requirements into policies requires careful reasoning about permissions, constraints, and exceptions, and
Thanks to tcclaviger, vLLM now has expert RAM offloading support (link). This makes frontier models much more accessible on a local setup! I was able to run the original DeepSeek-V4-Flash-Vision-Exp o
arXiv:2609.32819v1 Announce Type: cross Abstract: Raising the update-to-data (UTD) ratio breaks off-policy critics in two ways grouped as 'plasticity loss': collapsing representations and growing valu
arXiv:2609.31795v1 Announce Type: new Abstract: We describe our entry to the LoViF 2026 AIGC Image Compression Challenge, a benchmark for ultra-low-bitrate coding of AI-generated images under a strict
arXiv:2609.34367v1 Announce Type: new Abstract: Learned image codecs (LICs) achieve high reconstruction quality, but their decoding speed is often insufficient for immersive virtual reality (VR). Gaus
arXiv:2609.34170v1 Announce Type: new Abstract: Flow-based vision-language-action (VLA) models are highly effective for generalist robot manipulation, yet their reliance on computationally expensive V
arXiv:2609.32681v1 Announce Type: new Abstract: 4D radar provides geometric and motion cues that complement visual semantics, but integrating it into vision-language-action (VLA) models requires both
arXiv:2609.32416v1 Announce Type: cross Abstract: Code-as-Policy agents accomplish long-horizon embodied tasks by generating and executing code, yet continually improving them with teachers that are s
arXiv:2609.32802v1 Announce Type: cross Abstract: One variable sets what an upstream fault costs a staged pipeline of language-model agents: re-derivability, how much of what a stage needs it can rebu
arXiv:2609.33674v1 Announce Type: new Abstract: Graph neural networks (GNNs) are often called long-range because their architecture can connect distant nodes, but this does not show whether they use d
arXiv:2609.33065v1 Announce Type: cross Abstract: Large language model (LLM) assistants can now search the web and consult external sources while completing user requests. These sources can provide us
arXiv:2609.32757v1 Announce Type: cross Abstract: VLM bounding-box localization is both language generation and spatial commitment. Parseable fields such as bbox_2d make localization easy to score, bu
arXiv:2609.33202v1 Announce Type: new Abstract: Flow-matching models generate high-quality images and videos, but repeated neural network evaluations make sampling expensive. Skipping evaluations redu
arXiv:2609.32734v1 Announce Type: new Abstract: AI-generated image detectors are often evaluated on benchmarks where real and synthetic images differ in content, quality, or generation artifacts, allo
arXiv:2609.33095v1 Announce Type: new Abstract: Generating responsive listener facial motion is an important task for embodied conversational AI. Two modeling challenges are central: accounting for th
arXiv:2609.32141v1 Announce Type: new Abstract: Deep ReLU networks are piecewise-affine mappings that partition the input space into cells, each characterized by a distinct activation pattern. This st
arXiv:2609.32595v1 Announce Type: new Abstract: Safe and robust robot navigation across diverse environments requires a high-level understanding of complex scenes and the ability to carry it into stab
arXiv:2609.32155v1 Announce Type: new Abstract: Sequential manipulation requires a robot to track what has already happened, even when the current scene no longer reveals it. Policies with explicit hi
arXiv:2609.35200v1 Announce Type: cross Abstract: Memory-dependent manipulation requires robots to make decisions using information that is no longer available to their current sensors, such as recall
arXiv:2609.34567v1 Announce Type: new Abstract: This review examines the evolution of plant phenotyping from conventional manual trait measurement to high-throughput, robotic, and artificial intellige
arXiv:2505.09274v2 Announce Type: replace Abstract: Medical imaging is a cornerstone of modern healthcare, driving advancements in diagnosis, treatment planning, and patient care. Among its various ta
arXiv:2609.31927v1 Announce Type: cross Abstract: MathNet-Retrieve asks a retriever to find, for a math problem, a document stating the same problem. An LLM under one fixed prompt writes each gold doc
arXiv:2609.35075v1 Announce Type: new Abstract: Many sensing tasks obtain training labels only by deploying instruments in the field. With a limited number of sensor kits, a collection deadline, and m
arXiv:2609.34032v1 Announce Type: new Abstract: A manga reader meets a character on one page and knows them on sight a hundred pages later, without ever being handed a cast list. Re-identifying comic
arXiv:2609.35034v1 Announce Type: cross Abstract: Off-policy evaluation, which estimates evaluation policy performance from logged data, is key for recommender ranking policies. However, logged clicks
arXiv:2609.33420v1 Announce Type: cross Abstract: Recovering probability measures from their moments has numerous applications, esp. in connection with the method of moments in statistics and optimiza
arXiv:2609.32409v1 Announce Type: new Abstract: Amortized neural likelihoods enable computationally expensive inference for models with analytically intractable or unspecified likelihoods, but their b
arXiv:2609.33378v1 Announce Type: cross Abstract: A central goal in robotics is to enable manipulation across changing tasks and environments. Vision-language-action (VLA) models provide broad manipul
arXiv:2609.34257v1 Announce Type: cross Abstract: Repeatedly training language models on their own generated data may create a synthetic-data feedback loop in which errors and distributional biases ar
arXiv:2606.18022v2 Announce Type: replace Abstract: Masked diffusion models (MDMs) generate sequences by iteratively refining a partially masked state and committing tokens in parallel. We introduce r
arXiv:2609.33854v1 Announce Type: new Abstract: Driving policies require capabilities of scene understanding and future evolution prediction. To achieve this goal, current end-to-end models typically