Sample-Adaptive Zeroth-Order Post-Training
arXiv:2610.12566v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization offers a memory-efficient alternative to backpropagation for fine-tuning large language models. However, existing ZO meth
Knowledge catalogue
arXiv:2610.12566v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization offers a memory-efficient alternative to backpropagation for fine-tuning large language models. However, existing ZO meth
arXiv:2610.12910v1 Announce Type: new Abstract: Visual world models predict the consequences of actions from images, but planning with them is typically done by optimizers such as CEM and MPPI. Howeve
arXiv:2504.05727v3 Announce Type: replace Abstract: Autonomous mobile robots must operate smoothly in human-populated indoor environments, where onboard-only sensing suffers from limited perception ra
arXiv:2610.13819v1 Announce Type: new Abstract: In clinical practice, experts often begin by converting raw inputs into a set of expert features before making a decision. While black-box models traine
arXiv:2610.13207v1 Announce Type: new Abstract: Fake news detection under low resource conditions is challenged by the scarcity of labeled data and the difficulty of capturing subtle deceptive pattern
arXiv:2610.13766v1 Announce Type: cross Abstract: Geospatial foundation models can support disaster damage assessment, but high accuracy on familiar disasters may hide shortcut learning: prediction fr
arXiv:2610.13157v1 Announce Type: new Abstract: Reinforcement learning (RL) for search agents often suffers from evidence fabrication, a failure mode where agents bypass external information (E_{ext})
arXiv:2610.13812v1 Announce Type: new Abstract: Flow-matching and diffusion models are increasingly used as priors for linear inverse problems, however, most methods are bespoke to one class of trajec
arXiv:2610.12723v1 Announce Type: new Abstract: Neural surrogates that forecast a physical field from sparse, noisy sensors read a window of nearby sensors and a history of past snapshots. Practitione
arXiv:2610.12599v1 Announce Type: new Abstract: Sparse computation can scale the parametric capacity of LLMs without proportionally increasing active computation, but pretraining sparse architectures
arXiv:2610.12532v1 Announce Type: cross Abstract: Generative creative interfaces routinely report the user's position within a space of possibilities. Such displays imply that the quantities shown wer
arXiv:2610.13896v1 Announce Type: new Abstract: Neural semantic parsers generate meaning representations token by token, yet well-formedness depends on references, scope, and graph relations that unfo
arXiv:2610.12939v1 Announce Type: cross Abstract: Decoder hidden states from different layers of large language models (LLMs) often encode complementary information. However, existing speculative deco
arXiv:2610.12651v1 Announce Type: new Abstract: LLM agents executing multi-step workflows fail in two fundamentally different ways: recoverable slips, which further attempts can repair, and knowledge
arXiv:2610.13253v1 Announce Type: cross Abstract: In-context-learning Tabular Foundation Models (TFMs) are uniquely suited to scenarios with limited resources or privacy constraints, as they solve new
arXiv:2610.12790v1 Announce Type: new Abstract: Reliable robotic manipulation requires maintaining a firm grasp while strictly avoiding both contact slip and structural damage to the object. Satisfyin
arXiv:2610.12921v1 Announce Type: new Abstract: Brain-computer interfaces decode neural activity, yet class-label supervision (CLS) leaves relationships among classes implicit. We propose Language as
arXiv:2606.19821v2 Announce Type: replace Abstract: Key Performance Measurement (KPM) forecasting is essential for proactive network management of 5G and next-generation telecom networks. However, exi
arXiv:2610.13221v1 Announce Type: new Abstract: Cancer diagnosis and treatment increasingly rely on multi-modal diagnostic technologies that generate high-dimensional datasets. Although these datasets
arXiv:2610.12981v1 Announce Type: cross Abstract: Visual interventions provide a training-free way to reduce hallucinations in multimodal large language models. However, an intervention may have littl
arXiv:2610.13398v1 Announce Type: new Abstract: Tibetan, spoken by some seven million people on and around the Qinghai-Xizang Plateau, remains underrepresented in natural language processing (NLP). We
arXiv:2610.13810v1 Announce Type: cross Abstract: Large language models (LLMs) are now a standard source of query expansion terms but they hallucinate. They inject non existent entities and off-topic
arXiv:2610.12747v1 Announce Type: cross Abstract: Ego-trajectory prediction in urban roadway work zones remains an underexplored topic in autonomous-driving research, with limited dedicated data and s
arXiv:2610.13690v1 Announce Type: new Abstract: Diffusion models typically learn the generative dynamics of a single target distribution or a family of distributions indexed by conditioning variables
If you build on open-weight reasoning models, you have probably wanted a 'how sure is the model about this answer?' signal that does not cost 5x sampling, a judge model, or a labeled dataset. Hence, w
arXiv:2610.13783v1 Announce Type: new Abstract: Memes play a central role in modern online communication, especially in contexts of political conflict where visual and textual cues combine to convey p
arXiv:2412.06534v5 Announce Type: replace-cross Abstract: Understanding the internal mechanisms of deep neural networks remains a central challenge in machine learning. In computer vision, one promisi
arXiv:2610.13695v1 Announce Type: new Abstract: Any automation within reliability-critical contexts such as medicine or law has to be transparent and verifiable. Despite their advantages, current LLM-
arXiv:2610.12761v1 Announce Type: new Abstract: Foundation models are increasingly used for high-level multi-robot task planning, yet independently generated symbolic plans can become jointly infeasib
arXiv:2610.13145v1 Announce Type: new Abstract: Large language models (LLMs) increasingly generate structured layouts and executable scene programs from natural-language instructions. However, executa
arXiv:2610.13608v1 Announce Type: cross Abstract: Vocal intensity conversion renders a recorded utterance at a different vocal intensity, rather than simply scaling its amplitude, while preserving its
arXiv:2610.12995v1 Announce Type: cross Abstract: Supervised fine-tuning (SFT) remains a widely used approach for post-training large language models, because fixed demonstrations can directly introdu
arXiv:2610.13024v1 Announce Type: cross Abstract: World Action Models (WAMs) inherit powerful priors from pretrained video generation models, but their variational autoencoder (VAE) representations ar
arXiv:2610.12823v1 Announce Type: new Abstract: Web agents increasingly browse, interpret, and act on web content on behalf of users. In many realistic workflows, however, the visual evidence they con
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
arXiv:2610.12876v1 Announce Type: cross Abstract: Mixed-integer programming (MIP) solvers are widely used, yet their performance can vary substantially even under seemingly neutral perturbations. Such
arXiv:2610.13123v1 Announce Type: cross Abstract: A solid understanding of the predictive success of deep neural networks (DNNs) remains elusive. For example, while many data-dependent metrics exist t
arXiv:2610.12984v1 Announce Type: new Abstract: Computer-use agents can perform increasingly complex GUI tasks, yet unfamiliar software and workflows requiring human domain knowledge remain challengin
For the past few weeks I've been trying out converting existing dense models to sparse Mixture-of-Experts models, with no pretraining from scratch. The Idea If you can turn a dense model into an MoE t
Decision 3.0 now on Core ML 🍎 d3-lite triaging 1,000 pull requests into vllm-sr workgroups, live on an M5 Pro: 0.12 s per PR, 8.3 PRs/s, nothing leaves the Mac ⚡ lite, nano and mini are all converted
It's been 4 years since c.ai first hallucination model, and yet, we're nowhere good enough at LLMs in terms of spontaneity/interesting hallucination features. Fortunately, last year brought new archit
DISCLAIMER: the post and the model card was made with the assist of AI. Hiya, I’m releasing Recursive BitNet N-Gram 102M, a small experiment combining ternary weights, shared transformer layers, and h
Banger paper from Meta on AI research agents. (bookmark it) If you build AI-scientist systems, this one is worth your time. The result: A 27B open model beats the strongest open autoresearch baseline
Nice paper on agent memory. It explores an effective routing mechanism for memory in agents. The overall finding is that you want a separate retrieval strategy for each type of memory instead of one e
Reasoning From Scratch: Reinforcement Learning with Verifiable Rewards (RLVR) round 2. Covering clipped policy ratios, KL loss term, format rewards, and other GRPO tips & tricks. 00:00 Introduction an
arXiv:2607.17575v3 Announce Type: replace Abstract: We propose ARBITER, a novel LLM guardrail framework that introduces two key ideas: (i) dual-hypothesis reasoning, a reasoning method for LLM guardra
arXiv:2610.12113v1 Announce Type: cross Abstract: Off-policy actor--critic methods control overestimation bias by taking the minimum of two critics. This uses the same aggregation rule everywhere, reg
arXiv:2610.11022v1 Announce Type: cross Abstract: Skillful prediction of biomass-burning activity several days in advance is important for air-quality forecasting and aerosol prediction. Two operation
arXiv:2610.12420v1 Announce Type: new Abstract: Reinforcement learning in safety-critical domains requires maximizing task performance while strictly adhering to safety constraints. Existing safe rein
arXiv:2610.11585v1 Announce Type: cross Abstract: Recent advances in large language model (LLM) pretraining highlight the role of high-quality training data in improving performance. While model-based
arXiv:2610.10590v1 Announce Type: new Abstract: Tool-using agents repeatedly carry observations whose useful content can be much smaller than their original payload. We study agent-controlled forgetti
arXiv:2603.08706v2 Announce Type: replace Abstract: Imitation learning (IL) teaches language-model agents to reproduce expert actions but not to distinguish them from plausible mistakes. Self-reflecti
arXiv:2610.09944v2 Announce Type: replace Abstract: An essential control of AI agents is their ability to manage runtime. This ability requires a sense of time-awareness, to predict and estimate wall-
arXiv:2604.03672v2 Announce Type: replace-cross Abstract: Government agencies must register, classify and route every citizen appeal within statutory time limits, and much of this work is still done b
arXiv:2610.11162v1 Announce Type: new Abstract: Large pretrained clinical models provide a practical way to reuse learned prior knowledge across hospitals by adapting models to them. In practice, a ta
arXiv:2610.11128v1 Announce Type: new Abstract: A verified reference solution provides a correct trajectory for training a reasoning model. Alternatively, a prefix of the reference can guide the model
arXiv:2610.10663v1 Announce Type: cross Abstract: Bioacoustics and ethology encompass a wide range of audio understanding tasks, many of which stand to benefit from recent advances in large audio-lang
arXiv:2511.04655v2 Announce Type: replace Abstract: Multimodal LLMs can answer many questions in vision-centric benchmarks without looking at the image, using linguistic priors and skewed answer distr
arXiv:2610.11407v1 Announce Type: cross Abstract: Synthetic tabular generators are commonly optimized for distributional fidelity, but statistical similarity alone does not guarantee preservation of c
arXiv:2603.23406v3 Announce Type: replace Abstract: Generative agent societies simulate people with assigned roles, preferences and relationships. As agents exchange arguments and choose partners, the