43 articles mentioning Step-5-Preview-BF16
arXiv:2504.19024v2 Announce Type: replace Abstract: We propose a novel k-step return estimation method (called KETCHUP) for Reinforcement Learning(RL)-based knowledge distillation (KD) in text generation tasks. Our idea is to induce a K-step return by using the Bellman Optimality Equation for multip
arXiv:2507.11228v2 Announce Type: replace Abstract: Gradient descent on logistic regression has many fascinating properties not fully captured by classical analyses. For linearly separable data, the iterates converge in direction to the maximum-margin separator regardless of how large the step size
arXiv:2511.14102v2 Announce Type: replace Abstract: Mixture-of-experts (MoE) models often exceed a single accelerator's memory capacity. Offloading experts to host memory creates a severe PCIe I/O bottleneck because data-dependent routing keeps transfers on the critical path. We present Specter, an
arXiv:2610.27379v1 Announce Type: new Abstract: Block-diffusion speculative decoding accelerates large language model inference by drafting multiple tokens in parallel. However, one-step drafting in discrete diffusion is limited by token factorization error: each token is predicted from the same mas
arXiv:2610.27747v1 Announce Type: new Abstract: Muon and related optimizers update a weight matrix with a spectrally transformed gradient. In this work, we study such spectral optimizers with a constant step size on strongly convex matrix quadratics, which isolate the effect of the update from the c
arXiv:2610.06617v2 Announce Type: replace-cross Abstract: World Action Models (WAMs) incorporate visual representations from video generation backbones to guide action prediction. Recent efficient WAMs adopt Mixture-of-Transformers (MoT) architectures and compute video representations once for reuse
arXiv:2608.13179v2 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) directly optimizes verifier-defined outcomes for multi-turn tool-use agents, yet its trajectory-level credit assignment conflates heterogeneous per-turn outcomes into a single reward signal. On-
arXiv:2610.23284v1 Announce Type: new Abstract: Every study of why LLM agents fail begins by labelling traces by hand, and every such study builds its own throwaway tool. Observability platforms record agent traces but offer no structured labelling; annotation tools label text but have no notion of
arXiv:2610.23707v1 Announce Type: new Abstract: A multi-step agent task fails when a single tool it needs is missing, yet the tools an agent may need are too many to list: the roughly 8,200 tools on the platform we study have definitions that together exceed the context window of any model we evalua
arXiv:2607.18915v2 Announce Type: replace Abstract: With the rapid advancement of large language models (LLMs), modern systems not only possess strong foundational capabilities and extensive knowledge, but can also solve complex problems via long, multi-step reasoning. However, as reasoning traces b
arXiv:2610.21657v1 Announce Type: cross Abstract: Flow matching has emerged as a powerful generative family for simulation-based inference (SBI). However, sampling and density evaluation require numerical integration at inference time. This makes posterior sampling in neural posterior estimation cos
arXiv:2610.21125v1 Announce Type: new Abstract: Few-step generative models produce high-quality samples in one or a few evaluations. However, steering them toward desired properties remains difficult, especially when only black-box reward evaluations are available. While one could apply approaches d
arXiv:2607.04528v3 Announce Type: replace Abstract: Software agents are becoming increasingly important in modern coding workflows, where they solve complex tasks through multi-step interaction with code repositories. A central component of these systems is the execution harness, which governs how t
arXiv:2603.23463v3 Announce Type: replace-cross Abstract: Recent diffusion-based models achieve photorealism in image inpainting but require many sampling steps, limiting practical use. Few-step text-to-image models offer faster generation, but naively applying them to inpainting yields poor harmoni
arXiv:2610.21063v1 Announce Type: cross Abstract: AI agents now pause for approval before they send money, delete files or email outsiders, and that approval is usually treated as a second factor. We argue it is one only when compromising the agent cannot produce it and when it names exactly what it
arXiv:2610.20331v1 Announce Type: new Abstract: Diffusion and flow matching are now dominant frameworks for generative modeling in continuous domains, but they typically require multiple network evaluations at inference. This has motivated extensive research on one-step generative modeling primarily
arXiv:2610.19375v1 Announce Type: new Abstract: Latent world models are routinely selected, checkpointed and ablated on one-step prediction loss, on the assumption that a model which predicts the next latent better is a better model. In model-based RL this assumption is known to be unreliable; we as
arXiv:2610.19883v1 Announce Type: new Abstract: The growing interest in behavior foundation models, combined with recent advancements in generative modeling, has led to an appealing recipe for reinforcement learning from unlabeled data: pre-train a generative model to model occupancies, i.e., discou
arXiv:2610.13225v2 Announce Type: replace-cross Abstract: Generative modeling for discrete data has been significantly advanced by learning continuous vector fields on the statistical manifolds to transport simple priors to target discrete distributions. However, because these continuous-time flows
arXiv:2610.19601v1 Announce Type: new Abstract: Knowledge distillation (KD) improves language-model pretraining by matching teacher distributions at corpus prefixes, but provides no direct supervision at contexts induced by student predictions. We introduce PA1-OPD, which augments corpus-path distil
arXiv:2610.20071v1 Announce Type: new Abstract: MeanFlow predicts interval-average velocities, but conventional few-step sampling ties the interval queried from the model to the duration of the state update. We separate these two quantities at inference, fixing the model, update grid, and number of
arXiv:2610.19444v1 Announce Type: cross Abstract: Random reshuffling stochastic gradient descent (RR-SGD) is widely used for finite-sum optimization in machine learning. For convex $L$-smooth components, recent work establishes convergence guarantees for an averaged iterate with component stepsizes
arXiv:2509.24148v5 Announce Type: replace-cross Abstract: Test-Driven Development (TDD) is a widely adopted practice that requires developers to create and execute tests alongside implementation. With recent advances in Large Language Models (LLMs), developers can shift from manually writing the cod
arXiv:2610.20292v1 Announce Type: new Abstract: Reliable medical reasoning requires LLMs to dynamically access external biomedical evidence to improve factual trustworthiness and reasoning accuracy. However, indiscriminate retrieval can introduce noise and latency, raising a critical retrieval-timin
arXiv:2605.12805v2 Announce Type: replace-cross Abstract: MeanFlow enables one-step generation in continuous spaces by learning an average velocity over a time interval rather than the instantaneous velocity field of flow matching. However, discrete state spaces lack smooth trajectories or spatial d
arXiv:2610.12879v1 Announce Type: new Abstract: Diffusion language models generate tokens in parallel, but still require multiple sampling steps to produce high-quality text. In this paper, we investigate whether a model can learn to generate an entire sequence in a single forward pass. We propose S
arXiv:2610.13669v1 Announce Type: new Abstract: Curriculum learning typically organizes training samples from easy to hard. In GRPO, medium-difficulty samples are attractive for online selection because they are more likely to yield informative group-relative learning signals, and prioritizing them
arXiv:2605.21282v3 Announce Type: replace-cross Abstract: Combining entropy regularisation with mirror descent in online off-policy reinforcement learning requires expressive policies with efficient sampling and explicit entropy control. Gaussian policies struggle to represent multimodal targets, wh
arXiv:2607.04419v5 Announce Type: replace Abstract: Reusing evaluator-derived step rewards across evaluation channels raises a measurement question: does the assigned direction remain stable? We define Agent Step Value (ASV) as a channel-indexed before/after target-margin gain and identify the state
arXiv:2610.13105v1 Announce Type: new Abstract: Failure attribution for LLM agents aims to identify decisive steps responsible for task failure, which is critical for debugging and improving agentic systems. Existing methods rely on costly LLM prompting, step-level error annotations, or verified suc
arXiv:2610.12679v1 Announce Type: new Abstract: Efficient sampling from Bayesian posterior distributions in high dimensions remains a hard problem, despite advances in theory and algorithms. Multiple steps are usually required to draw unbiased samples, for example, iterations of an MCMC algorithm or
arXiv:2610.11334v1 Announce Type: new Abstract: In LLM-based agent systems, failures can originate from early steps whose effects propagate through subsequent interactions, making their origins difficult to identify. To trace such failures back to their origin, failure attribution has been formulate
arXiv:2609.06324v5 Announce Type: replace Abstract: Can a diffusion language model generate a coherent token block in one forward pass? Masked models already predict every position at once, but each prediction is the marginal distribution given the visible context, so the tokens can be mutually inco
arXiv:2610.12190v1 Announce Type: new Abstract: As claims about recursive self-improvement (RSI) and artificial general intelligence (AGI) proliferate, we ask a simple question: do frontier AI models have a sense of data, i.e., can they reliably select the right data for training? We introduce DataS
arXiv:2610.10859v1 Announce Type: cross Abstract: Text-to-image diffusion models are increasingly distilled into few-step variants and being deployed to enable fast inference. However, their ability to generate harmful or undesired content poses significant safety risks. Data-driven unlearning metho
arXiv:2610.12102v1 Announce Type: new Abstract: Flow matching has emerged as a scalable paradigm for training high-quality generative models, but sampling from the learned probability flow requires many network evaluations. Distillation can reduce this cost to one or a few evaluations; however, one-
arXiv:2610.11527v1 Announce Type: new Abstract: Atomistic simulation is a fundamental tool for studying long-term materials evolution, from diffusion and defect dynamics to interfacial reactions and fracture. Yet conventional simulators typically advance at microscopic resolution, spending substanti
arXiv:2610.10990v1 Announce Type: cross Abstract: Unified multimodal diffusion large language models (dLLMs) offer a single architecture for both image generation and multimodal understanding, but their iterative decoding requires tens to hundreds of forward passes. Existing few-step distillation me
arXiv:2607.12397v2 Announce Type: replace Abstract: LLM agents operate in stateful environments, where a single erroneous step can waste limited interaction budget or cause irreversible effects before task failure becomes apparent. Reliable deployment therefore requires step-level confidence estimat
arXiv:2606.00956v2 Announce Type: replace Abstract: This paper studies theoretical guarantees of a one-step lookahead Bayesian optimization (BO) method. Although the empirical effectiveness of one-step lookahead BO methods, such as entropy search, has been studied extensively, they often rely on com
arXiv:2610.09186v1 Announce Type: new Abstract: We argue that pattern recognition and step-by-step reasoning are two ends of a spectrum. A large language model (LLM) learns to reason step-by-step when data is structured such that the next token depends on a small amount of preceding context. Inferen
arXiv:2511.00053v2 Announce Type: replace-cross Abstract: The design of learning objectives is central to training time-series forecasting models. Existing learning objectives such as mean squared error mostly treat each future step as an independent, equally weighted task, which leads to the follow
arXiv:2610.07817v1 Announce Type: cross Abstract: Organizations automating operational processes need more than a correct outcome: they need to predict how a process will run, know which one actually ran, and inspect it step by step. When an agent is the executor that predictability is normally lost