Search: “models”

arXiv:2509.20909v2 Announce Type: replace Abstract: Large language models (LLMs) are commonly evaluated on challenging benchmarks such as AIME and Math500, where benchmark contamination can make memor

arXiv:2605.08697v1 Announce Type: new Abstract: The emerging collaborative information-based knowledge tracing (KT) has been a promising way to enhance modeling of learners' knowledge states. The core

arXiv:2605.08557v1 Announce Type: cross Abstract: Parameter-efficient adaptation of pretrained vision models is commonly performed through linear probes, prompts, low-rank updates, or lightweight resi

arXiv:2605.10067v1 Announce Type: cross Abstract: Red teaming is critical for uncovering vulnerabilities in Large Language Models (LLMs). While automated methods have improved scalability, existing ap

arXiv:2605.10616v1 Announce Type: cross Abstract: Tabular Foundation Models have recently established the state of the art in supervised tabular learning, by leveraging pretraining to learn generaliza

arXiv:2605.08503v1 Announce Type: new Abstract: Interactive narrative tasks require LLMs to sustain a coherent, evolving story while adapting to a user over multiple turns. However, suitable benchmark

arXiv:2605.08221v1 Announce Type: cross Abstract: This paper presents NoisyCoconut, a novel inference-time method that enhances large language model (LLM) reliability by manipulating internal represen

arXiv:2605.09996v1 Announce Type: new Abstract: While multimodal large language models have advanced across text, image, and audio, personalization research has remained primarily vision-language, wit

arXiv:2605.09822v1 Announce Type: cross Abstract: We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use p

arXiv:2605.10391v1 Announce Type: cross Abstract: We introduce Phoenix-VL 1.5 Medium, a 123B-parameter natively multimodal and multilingual foundation model, adapted to regional languages and the Sing

arXiv:2605.09228v1 Announce Type: cross Abstract: Most LLM benchmarks score how well a model responds to explicit requests. They leave unmeasured a different conversational ability: noticing and actin

arXiv:2509.26574v4 Announce Type: replace Abstract: While large language models (LLMs) with reasoning capabilities are progressing rapidly on high-school math competitions and coding, can they reason

arXiv:2512.20012v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are emerging as key enablers of automation in domains such as telecommunications, assisting with tasks including

arXiv:2605.08197v1 Announce Type: cross Abstract: Most causal benchmarks for language models score local answers or graph structure. We introduce ReplaySCM, a 1,300 item benchmark for executable causa

arXiv:2605.10831v1 Announce Type: cross Abstract: Large language models possess strong chemical reasoning capabilities, making them effective molecular editors. However, property-relevant information

arXiv:2605.09357v1 Announce Type: cross Abstract: Running deep neural networks on microcontroller units (MCUs) is severely constrained by limited memory resources. While TinyML techniques reduce model

arXiv:2605.08181v1 Announce Type: cross Abstract: Parameter-efficient fine-tuning methods introduce a small number of training parameters, enabling pre-trained models to adapt rapidly to new data dist

arXiv:2605.09844v1 Announce Type: new Abstract: The Metacognitive Probe is an exploratory five-task, 15-slot diagnostic that decomposes an LLM's confidence behaviour into five behaviourally-distinct d

arXiv:2511.02623v2 Announce Type: replace Abstract: Post-training alignment of large language models (LLMs) relies on large-scale human annotations guided by policy specifications that change over tim

arXiv:2605.09739v1 Announce Type: cross Abstract: Large Language Models are increasingly used as zero-shot classifiers in complex reasoning tasks. However, standard constrained decoding suffers from a

arXiv:2605.09544v1 Announce Type: new Abstract: Tool-integrated reasoning has emerged as a promising paradigm for enhancing large language models with external computation, retrieval, and execution ca

arXiv:2605.09734v1 Announce Type: cross Abstract: Most language-model training data shows final artifacts, not the process that produced them. We study a tractable version of this question in tool use

arXiv:2604.03701v2 Announce Type: replace Abstract: Video-based numerical reasoning provides a premier arena for testing whether Vision-Language Models (VLMs) truly 'understand' real-world dynamics, a

arXiv:2605.08344v1 Announce Type: new Abstract: Recent work has shown that models flow matching models can be trained without explicit time conditioning, challenging the standard view that the interpo

arXiv:2602.09789v3 Announce Type: replace Abstract: Scaling up model parameters has long been a prevalent training paradigm driven by the assumption that larger models yield superior generation capabi

arXiv:2605.08878v1 Announce Type: cross Abstract: Aligned large language models (LLMs) remain vulnerable to jailbreak attacks. Recent mechanistic studies have identified latent features and representa

arXiv:2605.08560v1 Announce Type: cross Abstract: We present ZAYA1-VL-8B, a compact mixture-of-experts vision-language model built upon our in-house language model, ZAYA1-8B. Despite its compact size,

arXiv:2605.07125v1 Announce Type: cross Abstract: Sequential recommendation has increasingly shifted toward generative recommenders that combine sequential patterns with semantic item information. Yet

arXiv:2605.07937v1 Announce Type: new Abstract: Long-horizon AI agents execute complex workflows spanning hundreds of sequential actions, yet a single wrong assumption early on can cascade into irreve

arXiv:2605.07562v1 Announce Type: new Abstract: Remote sensing vision-language models (RS-VLMs) face a fundamental mismatch with natural-image counterparts: the same geographic object exhibits radical

arXiv:2605.07985v1 Announce Type: cross Abstract: Selecting the optimal LLM inference configuration requires evaluation across hardware, serving engines, attention backends, and model architectures, s

arXiv:2605.07675v1 Announce Type: new Abstract: We introduce FactoryBench, a benchmark for evaluating time-series models and LLMs on machine understanding over industrial robotic telemetry. Q&A pairs

arXiv:2509.24789v4 Announce Type: replace Abstract: The evaluation of time series forecasting models is hindered by a lack of high-quality benchmarks, leading to overestimated assessments of progress.

arXiv:2602.14868v2 Announce Type: replace-cross Abstract: Reinforcement learning has emerged as a powerful paradigm for unlocking reasoning capabilities in language models. However, relying on sparse

arXiv:2605.07925v1 Announce Type: new Abstract: Conversational Large Language Models are post-trained on language that expresses specific behavioural traits, such as curiosity, open-mindedness, and em

arXiv:2605.05957v2 Announce Type: replace Abstract: LLMs reliably correct false claims when presented in isolation, yet when the same claims are embedded in task-oriented requests, they often comply r

arXiv:2605.05949v2 Announce Type: replace Abstract: Algorithmic problem solving serves as a rigorous testbed for evaluating structured reasoning in AI coding systems, as it directly reflects a model's

arXiv:2605.06903v1 Announce Type: cross Abstract: Large language models are now embedded in everyday writing workflows, making reliable AI-generated text detection important for academic integrity, co

arXiv:2605.07546v1 Announce Type: new Abstract: Neural scaling laws establish a predictable relationship between model performance and data or compute, offering crucial guidance for resource allocatio

arXiv:2605.07302v1 Announce Type: new Abstract: Finetuning pretrained models occurs in a low-dimensional subspace of the full parameter space. Prior work has focused on characterizing this optimizatio

arXiv:2605.07141v1 Announce Type: cross Abstract: Open-world referring segmentation requires grounding unconstrained language expressions to precise pixel-level regions. Existing multimodal large lang

arXiv:2605.05995v2 Announce Type: replace-cross Abstract: The safety alignment of Large Language Models (LLMs) remains vulnerable to Harmful Fine-tuning (HFT). While existing defenses impose constrain

arXiv:2603.05117v3 Announce Type: replace Abstract: Imitation Learning (IL) enables robots to acquire manipulation skills from expert demonstrations. Diffusion Policy (DP) models multi-modal expert be

arXiv:2504.16559v3 Announce Type: replace Abstract: Modeling the joint distribution of data samples and their properties allows to construct a single model for both data generation and property predic

arXiv:2601.21839v2 Announce Type: replace-cross Abstract: Test-time compute has emerged as a promising strategy to enhance the reasoning abilities of large language models (LLMs). However, this strate

arXiv:2605.07127v1 Announce Type: cross Abstract: Modern large language models (LLMs) can find a needle in a haystack (locating a single relevant fact buried among hundreds of thousands of irrelevant