Search: “inference”

arXiv:2604.19730v1 Announce Type: cross Abstract: Some of the most performant reinforcement learning algorithms today can be prohibitively expensive as they use test-time scaling methods such as sampl

arXiv:2604.18953v1 Announce Type: new Abstract: Deep learning surrogates for CFD flow-field prediction often rely on large, complex models, which can be slow and fragile when data are noisy or incompl

Flux 2-Klein-9B NVFP4 is a quantized image generation model that runs on mid-range GPUs like the RTX 3050. On this hardware, the model produces 1024-resolution images but with relatively slow inferenc

arXiv:2604.18873v1 Announce Type: new Abstract: Large language models (LLMs) are highly capable at language generation, but they remain unreliable when reasoning requires explicit symbolic structure,

arXiv:2602.15423v3 Announce Type: replace-cross Abstract: As the burgeoning power requirements of sophisticated neural architectures escalate, the information retrieval community has recognized ecolog

arXiv:2604.04516v2 Announce Type: replace-cross Abstract: Adapting LLMs to new domains causes forgetting because standard methods (e.g., full fine-tuning, LoRA) inject new directions into the weight s

arXiv:2604.18933v1 Announce Type: cross Abstract: Robotic manipulation tasks exhibit varying memory requirements, ranging from Markovian tasks that require no memory to non-Markovian tasks that depend

arXiv:2604.18804v1 Announce Type: cross Abstract: Latent Diffusion Models (LDMs) achieve high-fidelity synthesis but suffer from latent space brittleness, causing discontinuous semantic jumps during e

arXiv:2604.19624v1 Announce Type: new Abstract: Reconstructing physically plausible 3D human-scene interactions (HSI) from a single image currently presents a trade-off: optimization based methods off

arXiv:2604.19398v1 Announce Type: new Abstract: Large language models (LLMs) are expensive to serve because model parameters, attention computation, and KV caches impose substantial memory and latency

arXiv:2604.19669v1 Announce Type: new Abstract: Enforcing constraint satisfaction in neural network outputs is critical for safety, reliability, and physical fidelity in many control and decision-maki

Highlights: 👉 80.2% SWE-Bench Verified and 89.6% LiveCodeBench v6 👉 Agent Swarm executes up to 4,000 coordinated steps 👉 Native text, image, and video input with 79.4% MMMU-Pro 👉 Production-ready on t

arXiv:2604.19149v1 Announce Type: cross Abstract: Thinking LLMs produce reasoning traces before answering. Prior activation steering work mainly targets on shaping these traces. It remains less unders

How far are we from agents that can self-generate world knowledge? The work proposes an outcome-based reward that measures how much an agent's self-generated world knowledge actually improves its task

arXiv:2407.12027v2 Announce Type: replace-cross Abstract: In the rapidly evolving Internet of Things (IoT) domain, we concentrate on enhancing energy efficiency in Deep Learning accelerators on FPGA-b

arXiv:2604.19186v1 Announce Type: cross Abstract: Heterophily is a prevalent property of real-world graphs and is well known to impair the performance of homophilic Graph Neural Networks (GNNs). Prior

arXiv:2509.21080v2 Announce Type: replace-cross Abstract: Advancements in Large language models (LLMs) have enabled a variety of downstream applications like story and interview script generation. How

Introducing Kimi K2.6 from @Kimi_Moonshot, a multimodal agentic model with Agent Swarm scaling to 300 sub-agents and long-horizon coding stability. AI natives can now use Kimi K2.6 on Together AI and

Traditional data catalogs were built as manual inventories for technical users, focusing on table structures rather than the deep context that AI agents need. When agents lack business semantics and d

arXiv:2604.18729v1 Announce Type: new Abstract: Humor holds up a mirror to social perception: what we find funny often reflects who we are and how we judge others. When language models engage with hum

arXiv:2604.19680v1 Announce Type: new Abstract: In image restoration, single-step discriminative mappings often lack fine details via expectation learning, whereas generative paradigms suffer from ine

arXiv:2505.09803v3 Announce Type: replace-cross Abstract: In many applications, we wish to fit a parametric statistical model to a small ensemble of spatially distributed random variables ('fields').

arXiv:2604.19167v1 Announce Type: cross Abstract: Deploying large language models (LLMs) in resource-constrained environments is hindered by heavy computational and memory requirements. We present LBL

arXiv:2501.02200v2 Announce Type: replace-cross Abstract: The iterative search process of evolutionary algorithms (EAs) encapsulates optimization knowledge within historical populations and fitness ev

arXiv:2604.18878v1 Announce Type: new Abstract: We introduce LegalBench-BR, the first public benchmark for evaluating language models on Brazilian legal text classification. The dataset comprises 3,10

🦤 LeWorldModel: Learning Physics from Pixels — Stable World Models with Just Two Losses World models: 1️⃣ DINO-WM: pretrained ViT encoder (from ImageNet) → features → predictor. But encoder is frozen,

arXiv:2604.18940v1 Announce Type: new Abstract: Autonomous driving systems often degrade under adverse visibility conditions-such as rain, nighttime, or snow-where online scene geometry (e.g., lane di

arXiv:2506.09373v3 Announce Type: replace-cross Abstract: The advent of autonomous agents is transforming interactions with Graphical User Interfaces (GUIs) by employing natural language as a powerful

arXiv:2601.21092v2 Announce Type: replace Abstract: Planning effective interventions in biological systems requires treatment-effect models that adapt to unseen biological contexts by identifying thei

arXiv:2602.09642v2 Announce Type: replace-cross Abstract: Recent advances in Large Language Models (LLMs) have significantly improved table understanding tasks such as Table Question Answering (TableQ

arXiv:2604.18970v1 Announce Type: new Abstract: We can often verify the correctness of neural network outputs using ground truth labels, but we cannot reliably determine whether the output was produce

arXiv:2604.19675v1 Announce Type: new Abstract: Flow matching has recently emerged as a principled framework for learning continuous-time transport maps, enabling efficient deterministic generation wi

arXiv:2604.19642v1 Announce Type: new Abstract: Edge devices such as smartwatches and smart glasses cannot continuously run even the smallest 100M-1B parameter language models due to power and compute

arXiv:2602.12708v2 Announce Type: replace Abstract: Vertical Federated Learning (VFL) has emerged as a critical paradigm for collaborative model training in privacy-sensitive domains such as finance a

arXiv:2604.19093v1 Announce Type: cross Abstract: Multi-modal test-time adaptation (TTA) enhances the resilience of benchmark multi-modal models against distribution shifts by leveraging the unlabeled

arXiv:2502.16161v2 Announce Type: replace-cross Abstract: Visually-situated text parsing (VsTP) has recently seen notable advancements, driven by the growing demand for automated document understandin

arXiv:2602.01651v2 Announce Type: replace-cross Abstract: Why do neural networks fail to generalize addition from 16-digit to 32-digit numbers, while a child who learns the rule can apply it to arbitr

arXiv:2604.19656v1 Announce Type: new Abstract: Large language models have achieved remarkable progress on complex reasoning tasks. However, they often implicitly fabricate information when inputs are

arXiv:2403.09905v5 Announce Type: replace-cross Abstract: Embodied navigation methods commonly operate in static environments with stationary objects. In this work, we present approaches for tackling

arXiv:2604.19036v1 Announce Type: new Abstract: Defeasible statements are statements that are likely, or probable, or usually true, but may occasionally be false. Plausible reasoning makes conclusions

arXiv:2604.18649v1 Announce Type: cross Abstract: As generative AI faces intensifying legal challenges, the machine learning community has increasingly relied on post-hoc mitigation -- especially mach

arXiv:2604.19684v1 Announce Type: new Abstract: Explainable artificial intelligence (XAI) has predominantly focused on generating model-centric explanations that approximate the behavior of black-box

arXiv:2604.19083v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable success in cross-modal understanding and generation, yet their deployment is threate

arXiv:2506.06211v2 Announce Type: replace-cross Abstract: Puzzlehunts are a genre of complex, multi-step puzzles lacking well-defined problem definitions. In contrast to conventional reasoning benchma

arXiv:2604.00161v2 Announce Type: replace Abstract: Optical Character Recognition (OCR) is increasingly regarded as a foundational capability for modern vision-language models (VLMs), enabling them no

arXiv:2602.05416v2 Announce Type: replace-cross Abstract: While proper orthogonal decomposition (POD)-based surrogates are widely explored for hydrodynamic applications, the use of Koopman autoencoder