47 articles mentioning faster-whisper-large-v3
arXiv:2610.20682v1 Announce Type: cross Abstract: We introduce TurboMamba, a language-model architecture for faster inference, faster training, and better token prediction. A global Mamba-3 model runs once per completed token block. A small Transformer uses this global context and earlier tokens wit
arXiv:2610.21594v1 Announce Type: new Abstract: Looped Transformers improve parameter efficiency by repeatedly applying the same Transformer layers. However, generating each token still involves sequential computation and repeated access to model weights, leading to high decoding latency. Recent pip
Anant Kumar builds tools to compare loan valuations and analyze credit agreements, using Codex for code review and faster follow-ups.
arXiv:2508.16939v2 Announce Type: replace Abstract: Diffusion models achieve state-of-the-art generative performance but remain computationally expensive at inference, often requiring many discretization steps. We propose Sig-DEG (Signature-based Differential Equation Generator), a generator for dis
arXiv:2610.17948v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) have attracted increasing research attention due to their bidirectional attention and arbitrary order decoding. However, there are two issues in dLLMs inference: the decoding difficulty is dominated by the toke
arXiv:2610.14448v1 Announce Type: cross Abstract: Asynchronous rollout improves the efficiency of reinforcement learning with verifiable rewards (RLVR) by reducing under-utilization during long-tail generation. Sustained concurrency also concentrates context processing into fewer decode steps, incre
arXiv:2610.13092v1 Announce Type: new Abstract: AI agents depend on repeated decisions to accomplish tasks, with large language models (LLMs) commonly serving as their core decision makers. The recent emergence of Jev offers an opportunity to make these decisions faster and at lower cost. However, i
arXiv:2610.13835v1 Announce Type: new Abstract: An ideal (language) model must continually update its knowledge. Existing work largely focuses on making these updates efficient after training. We investigate how the organization of time-varying knowledge during pre-training shapes the efficiency of
arXiv:2610.10845v1 Announce Type: cross Abstract: A large language model can only use the text that fits in its context window, and it recomputes its internal key-value (KV) state for a prompt every time the prompt is sent. We test a memory layer, the public package galahad-kv, that saves the KV sta
arXiv:2610.04875v2 Announce Type: replace Abstract: Diffusion large language models (DLLMs) generate text through iterative block denoising, and multi-branch speculative decoding accelerates this process by verifying a main branch together with multiple draft branches in a single forward pass. While
arXiv:2608.02365v2 Announce Type: replace-cross Abstract: World Action Models (WAMs) build on pretrained video models, whose representations are grounded in physical dynamics and provide a natural basis for action prediction. Despite this natural foundation, many WAMs still rely on deep, parameter-h
arXiv:2610.04753v2 Announce Type: replace-cross Abstract: Autoregressive generation in Large Language Models (LLMs) is constrained by the memory and computational demands of attention mechanisms. Sparse attention methods mitigate this cost by selecting only high-probability entries of the attention
arXiv:2605.05660v3 Announce Type: replace Abstract: Multi-objective optimization (MOO) has received growing attention in applications that require learning under multiple criteria. However, most existing MOO formulations do not explicitly account for distributional shifts in the data. We introduce d
arXiv:2610.02598v1 Announce Type: new Abstract: Deploying LLMs on consumer-grade GPUs with insufficient memory to hold their weights can result in prohibitively slow inference, because decoding repeatedly transfers offloaded weights from system RAM or flash storage into GPU at much lower bandwidth t
arXiv:2610.01882v1 Announce Type: new Abstract: Multi-agent reinforcement learning (MARL) provides a powerful framework for learning coordinated behaviors through interactions with the environment. Developing MARL policies requires balancing expressive modeling of complex and multimodal action distr
arXiv:2609.32540v2 Announce Type: replace-cross Abstract: Few-step autoregressive video diffusion generates a long video by splitting the video into temporal chunks and generating chunk-by-chunk, each through a short sequence of denoising stages. To memorize chunks that are already generated, previo
arXiv:2609.01730v2 Announce Type: replace-cross Abstract: Fully homomorphic encryption (FHE) allows a server to run a language model directly on encrypted user prompts, but current approaches remain prohibitively slow. Ciphertexts natively support only addition, multiplication, and rotation, and mul
arXiv:2609.36610v1 Announce Type: new Abstract: Agnostic federated learning (AFL) seeks a model that performs reliably across $m$ heterogeneous workers, but communication remains a bottleneck. We improve communication efficiency by reducing the number of synchronization rounds via faster convergence
arXiv:2609.29769v2 Announce Type: replace Abstract: LLM judges score outputs against rubrics well enough to have become the norm, both in benchmarks and as rewards for training. Jev, a classifier-like alternative its creators call a "decision model", returns probabilities over permitted answers with
arXiv:2609.30474v1 Announce Type: new Abstract: Speculative decoding is a successful technique speeding up inference of a target autoregressive language model via a fast drafter model. Lossy speculative decoding allows a drift with respect to the target to further improve speed. Interestingly, it ha
Anthropic has released the newest version of its mid-range model, boasting faster response times and less token burn.
GPT-6 Astra completed a 50-tab tax workbook twice as fast as GPT-5.6 Sol, and its stronger understanding of user intent gives Basis more confidence in real-world use.
arXiv:2609.28947v1 Announce Type: cross Abstract: Misinformation is widely reported to propagate faster on engagement-based platforms, yet prior work largely focused on empirical analysis, without identifying a specific algorithmic mechanism that results in this phenomenon. Thanks to the open-sourci
arXiv:2609.14146v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) deployment can reduce inference latency while changing closed-loop task behavior. We evaluate HuggingFaceVLA/smolvla_libero on an RTX 2060 (6 GB) in LIBERO Spatial and Object (MuJoCo 3.3.2, LeRobot 0.6.1, seed 42)
arXiv:2602.11139v2 Announce Type: replace Abstract: Tabular foundation models, such as TabPFNv2 and TabICL, have recently dethroned gradient-boosted trees at the top of predictive benchmarks, demonstrating the value of in-context learning for tabular data. We introduce TabICLv2, a new state-of-the-a
arXiv:2609.15177v1 Announce Type: new Abstract: Diffusion language models (dLLMs) promise fast inference by generating multiple tokens in parallel, but suffer severe performance degradation when parallel decoding is pushed too aggressively. We introduce Temporal Self-Distillation (TSD), a simple on-
arXiv:2609.11923v1 Announce Type: cross Abstract: Counterfactual regret minimization (CFR) is one of the few large numerical workloads that still runs faster on CPUs than on GPUs. Each iteration sweeps a game tree with up to billions of states in millions of small, interdependent gather and scatter
arXiv:2609.07762v1 Announce Type: cross Abstract: Recurrent detectors such as bidirectional long short-term memory (Bi-LSTM) networks are low-complexity alternatives to the optimal Bahl-Cocke-Jelinek-Raviv (BCJR) detector for faster-than-Nyquist (FTN) signaling. Motivated by convolutional detectors
arXiv:2605.20314v2 Announce Type: replace-cross Abstract: This work investigates the ``small-vs-large gap'', where repeating on fewer samples can lead to compute saving during training compared to using a larger dataset. This is observed across algorithmic tasks, architectures and optimizers and can
arXiv:2609.03236v1 Announce Type: new Abstract: Tool-using LLM agents spend wall-clock time not only on model inference but also in serial action--observation turns, where each tool call, environment transition, and observation can delay subsequent decisions. We introduce \textbf{Speculative Macro C
arXiv:2609.00097v1 Announce Type: cross Abstract: The development of long-context Large Language Models (LLMs) is constrained by the memory bandwidth bottleneck and quadratic complexity of the attention mechanism during decoding. To overcome the inherent trade-offs between the memory overhead of met
Elon Musk says a secretive new SpaceX foundry will let him cast his own turbine blades and get gas power online 18 months faster than anyone else — but it's a bet on a fuel source that's already triggering lawsuits and health studies everywhere his (and others') turbines have gone in.
arXiv:2504.07742v4 Announce Type: replace-cross Abstract: Bayesian optimization (BO) is an effective technique for black-box optimization. However, its applicability is typically limited to moderate-budget problems due to the cubic complexity of fitting the Gaussian process (GP) surrogate model. In
arXiv:2608.25903v1 Announce Type: cross Abstract: Relational Deep Learning (RDL) is an effective approach to machine learning over multi-table relational databases. In RDL, a database is modeled as a graph in which each row is a node and each foreign-key relation is an edge, and a graph neural netwo
arXiv:2607.29079v1 Announce Type: new Abstract: Training-free acceleration makes diffusion-based multimodal large language models (dMLLMs) more deployable, but it may silently change generated content. We study this serving-time consistency problem on 300 real images, comparing Fast-dLLM outputs wit
arXiv:2607.27842v1 Announce Type: cross Abstract: Diffusion models are widely used to generate high-quality images and videos, but their iterative denoising process remains computationally intensive. A growing class of training-free accelerators reduces this cost by reusing cached intermediate featu
arXiv:2602.14649v2 Announce Type: replace Abstract: Large Language Models (LLMs) exhibit strong reasoning abilities, but their high computational costs limit their practical deployment. Recent studies reveal significant redundancy in LLMs layers, making layer pruning an active research topic. Layer
arXiv:2607.18292v3 Announce Type: replace-cross Abstract: Bigger language models are less reliable. Across three families, three benchmarks and six rungs, including in-the-wild chat logs, scaling closes the start-of-response knowledge gap up to $7\times$ while within-response knowledge degradation g
arXiv:2605.12183v2 Announce Type: replace Abstract: Drifting Models have emerged as a new paradigm for one-step generative modeling, achieving strong image quality without iterative inference. The premise is to replace the iterative denoising process in diffusion models with a single evaluation of a
arXiv:2607.18292v2 Announce Type: replace-cross Abstract: As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability. The knowledge-gap account -- more data, retrieval, or scale -- misses an auto-regressive risk residual that increases with scale
arXiv:2607.21042v1 Announce Type: new Abstract: Autoregressive text-to-speech models achieve strong naturalness but suffer from slow inference due to sequential token generation, limiting their deployment in production applications that require low latency. IndexTTS-2 is a state-of-the-art autoregre
arXiv:2607.18292v1 Announce Type: cross Abstract: As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability. The knowledge-gap account - more data, retrieval, or scale - misses an auto-regressive risk residual that scale sharpens: the model comm
arXiv:2607.15745v1 Announce Type: new Abstract: Common practice when training Convolutional Neural Networks (CNNs) is to use randomly shuffled mini-batches. This creates two limitations: slower convergence, and a diminishing learning signal, since many samples are quickly classified as easy during t
arXiv:2607.11624v3 Announce Type: replace-cross Abstract: Reinforcement learning (RL) algorithms classically suffer from poor sample efficiency. In robotics, a recent line of work has emerged addressing this problem by encoding physics priors in the learning process. However, most of these approache
Across 107 enterprises, AI infrastructure spending is accelerating well ahead of the ability to see or steer its economics. Most organizations run their AI on a familiar base of hyperscalers and model-provider APIs, yet the next dollar is aimed at specialized compute almost none of them use today; a
arXiv:2607.12696v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) models have become an important approach for scaling Large Language Models (LLMs), but their inference efficiency depends strongly on expert activation patterns. Speculative decoding (SD) accelerates autoregressive gener