Knowledge catalogue
arXiv:2608.04934v1 Announce Type: new Abstract: Training LLM agents commonly relies on supervised fine-tuning from expert trajectories or online reinforcement learning over human-specified tasks with
arXiv:2608.04904v1 Announce Type: new Abstract: Multilingual large language models exhibit substantial performance differences across languages, while existing adaptation methods often require paramet
arXiv:2608.04567v1 Announce Type: new Abstract: Event knowledge concerns who does what to whom. Psycholinguists use event-plausibility judgments to examine how this knowledge supports human language p
arXiv:2608.04183v1 Announce Type: new Abstract: When a language model follows an in-context conditional rule such as 'if P(x) then A else B,' does it assemble a runtime circuit with one module that te
arXiv:2608.04355v1 Announce Type: new Abstract: Accuracy changes after language-model self-revision are usually interpreted as changes in reasoning. We show this can fail at the answer-extraction boun
arXiv:2608.04463v1 Announce Type: new Abstract: Prior work on LLM conformity largely measures discrete answer flips under verifiable labels. Open-ended revisions require a different measurement strate
arXiv:2608.04268v1 Announce Type: new Abstract: Generative models trained on artificially generated data have been shown to exhibit model collapse, resulting in significant performance degradation. As
arXiv:2608.04570v1 Announce Type: new Abstract: Personalized LLMs with persistent memory are increasingly deployed, yet the faithfulness of their user models remains unexamined. We study over-inferenc
arXiv:2603.18409v2 Announce Type: replace Abstract: Current document chunking methods for Retrieval-Augmented Generation (RAG) typically linearize text. This forced linearization strips away intrinsic
arXiv:2601.22124v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly adapted for medical applications, but most are trained using data from a single institution because pr
arXiv:2608.05139v1 Announce Type: new Abstract: Long-horizon reasoning in recent LLMs demands that the model switch between distinct skills inside a reasoning chain, such as first doing a math derivat
arXiv:2608.04260v1 Announce Type: new Abstract: Metaphorical language remains a major challenge for multilingual natural language processing because successful interpretation and translation require r
arXiv:2608.04759v1 Announce Type: cross Abstract: Although Multimodal Large Language Models (MLLMs) have made substantial progress, their spatial reasoning may still produce intermediate judgments inc
arXiv:2608.04015v1 Announce Type: new Abstract: With the increase in digitized resources of Classical Latin texts and modern breakthroughs of Large Language Models (LLMs), I contribute to ancient lang
arXiv:2607.19322v2 Announce Type: replace Abstract: Rubric-based evaluation of open-ended generation faces a fundamental tension between expressiveness and reliability. Authoring a faithful rubric req
arXiv:2608.04949v1 Announce Type: cross Abstract: Unified Multimodal Relation Extraction (UMRE) aims to identify intra-modal and cross-modal relations between textual entities and visual objects. Howe
arXiv:2409.02228v2 Announce Type: replace-cross Abstract: When language models (LMs) are trained to forget (or 'unlearn'') a skill, how precisely does their behavior change? We study the behavior of t
arXiv:2608.04574v1 Announce Type: new Abstract: Memory-augmented VLM agents act on persistent spatial knowledge, yet that knowledge silently goes stale as the environment changes. We ask what happens
arXiv:2604.02778v2 Announce Type: replace Abstract: Real-world multimodal knowledge graphs (MMKGs) are dynamic, with new entities, relations, and multimodal knowledge emerging over time. Existing cont
arXiv:2608.04021v1 Announce Type: new Abstract: Cloze-style probes that vary how often a target token appears implicitly assume that more copies of a target affect prediction the same way regardless o
arXiv:2608.03617v1 Announce Type: new Abstract: The personal archive of Konstantin Tsiolkovsky (1857-1935) is held as fond 555 of the Archive of the Russian Academy of Sciences. The archive scanned th
arXiv:2608.02989v1 Announce Type: cross Abstract: Speculative decoding verifies a tree of draft tokens in one target-model forward pass. For a mixture-of-experts (MoE) target, however, parallel verifi
arXiv:2608.03067v1 Announce Type: new Abstract: Changes in spontaneous speech provide an early signal of cognitive dysfunction in Alzheimer's disease (AD) that large language models (LLMs) can detect.
arXiv:2608.03999v1 Announce Type: cross Abstract: Text-to-music language models begin with a choice usually made by default: how to tokenize music. Normally entangled with backbone, data, and recipe,
arXiv:2608.02941v1 Announce Type: new Abstract: We audit five frontier large language models on native Bangla derogatory speech (gali) across six protocols to test a single hypothesis: Comprehension-C
arXiv:2608.03735v1 Announce Type: cross Abstract: Multilingual multi-agent systems exhibit substantial degradation beyond English, yet prior work rarely identifies how task-critical information is los
arXiv:2608.03154v1 Announce Type: new Abstract: Biomedical relation extraction (BioRE) extracts structured knowledge from biomedical literature for applications such as knowledge base construction and
arXiv:2608.02901v1 Announce Type: cross Abstract: The key-value (KV) cache is the primary memory bottleneck in long-context LLM inference. Existing approaches attack it from opposite ends: eviction me
arXiv:2608.03898v1 Announce Type: new Abstract: The automatic structural analysis of legal texts is a cornerstone of legal technology, yet the extraction of their logical components remains a signific
arXiv:2608.02703v1 Announce Type: new Abstract: Weight-only quantization substantially reduces the storage of large language model (LLM) transformer blocks, but practical backends often retain the fin
arXiv:2608.03358v1 Announce Type: new Abstract: Existing visual emotion understanding methods typically ignore cultural variations in emotional perception. We introduce culture-conditioned visual emot
arXiv:2608.02947v1 Announce Type: cross Abstract: The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelen
arXiv:2608.03711v1 Announce Type: cross Abstract: In human visual perception, uppercase lettering serves as a natural salience cue that captures attention within lowercase text. In this paper, we pres
arXiv:2608.03884v1 Announce Type: cross Abstract: In-the-wild Bengali scene text recognition is largely unmeasured: existing resources target handwritten documents or constrained sign-board parsing, r
arXiv:2608.02612v1 Announce Type: new Abstract: Formulating an optimization problem strongly affects the quality of the final solution, yet good formulations usually require substantial expertise. Rec
arXiv:2608.03340v1 Announce Type: new Abstract: Predicting LLM's capabilities on real-world tasks is essential, yet the extent to which performance on commonsense benchmarks predicts downstream perfor
arXiv:2608.03038v1 Announce Type: new Abstract: Statistical reasoning is multidimensional, yet evaluations of large language models (LLMs) typically emphasize response accuracy while overlooking how m
arXiv:2608.03494v1 Announce Type: new Abstract: Vocabulary extension is an efficient way to adapt pretrained large language models (LLMs) to new languages, but the initialization of newly added token
arXiv:2601.17003v2 Announce Type: replace-cross Abstract: Mental-health AI safety is typically evaluated with small, simulation-based benchmarks that may not reflect the linguistic and contextual dive
arXiv:2506.13502v3 Announce Type: replace Abstract: Next-word prediction (NWP) trains language models against a single observed continuation, even though many contexts admit multiple plausible next wo
arXiv:2607.17447v2 Announce Type: replace-cross Abstract: As generative artificial intelligence enters scientific and professional work, its uncertainty must be defined on the states that matter for i
arXiv:2608.02935v1 Announce Type: new Abstract: Arabic script uses 28 letters, many of which share a common base shape (rasm) and are distinguished only by dot placement. Because early Arabic manuscri
arXiv:2608.03247v1 Announce Type: cross Abstract: Multimodal learning has significantly advanced survival prediction by integrating pathology images with genomic data. However, clinical information, d
arXiv:2608.03505v1 Announce Type: new Abstract: Constructed languages (conlangs) are intentionally created human languages with a rich tradition of linguistic creativity. Despite their potential for s
arXiv:2608.03529v1 Announce Type: new Abstract: Biomedical literature is increasingly mined for knowledge beyond the questions it was written to answer. Because the target concepts are not known in ad
arXiv:2608.03108v1 Announce Type: cross Abstract: Offline reinforcement learning (offline RL) can benefit from nearby out-of-distribution (OOD) actions, but estimation errors at these actions may be a
arXiv:2608.02694v1 Announce Type: new Abstract: Long-horizon video editing agents receive final-product feedback only after many interdependent decisions. Yet editing quality is subjective, admits mul
arXiv:2608.03532v1 Announce Type: new Abstract: Large language models are increasingly deployed in multilingual contexts, yet safety alignment and bias evaluation remain overwhelmingly English-centric
arXiv:2608.03720v1 Announce Type: new Abstract: Large language models can generate fluent responses to Islamic questions while introducing factual errors that are difficult to identify. This paper pre
arXiv:2608.03599v1 Announce Type: new Abstract: Byte-level language models are usually argued for on the grounds of robustness, multilingual fairness, and character-level skills. We point to a differe
arXiv:2604.06005v2 Announce Type: replace Abstract: Interpreting the information encoded in language model weights remains a fundamental challenge in mechanistic interpretability. In this work, we int
arXiv:2608.03388v1 Announce Type: new Abstract: Abductive reasoning requires forming hypotheses that explain observed evidence and revising them as new evidence becomes available. While large language
arXiv:2510.11834v3 Announce Type: replace-cross Abstract: Generative models are increasingly paired with safety classifiers that filter harmful or undesirable outputs. A common strategy is to fine-tun
arXiv:2608.03130v1 Announce Type: cross Abstract: Long-term memory enables persistent personalization in LLM agents, but repeated memory-conditioned responses can cumulatively reveal protected attribu
arXiv:2608.03883v1 Announce Type: new Abstract: This paper describes our submissions to eRisk 2026 Task 3, ADHD Symptom Sentence Ranking. The task requires systems to rank candidate Reddit sentences a
arXiv:2608.03411v1 Announce Type: new Abstract: Accurate Uncertainty Quantification (UQ) is critical for reliable deployment of Large Language Models (LLMs), yet traditional probability-based metrics
arXiv:2608.03437v1 Announce Type: new Abstract: While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability. When identifying top-performing m
arXiv:2608.03480v1 Announce Type: new Abstract: The adoption of large pre-trained multilingual models for neural machine translation (MNMT) faces a major challenge: excessive memory and computational
arXiv:2608.02966v1 Announce Type: new Abstract: Most multiple-choice question (MCQ) benchmarks evaluate Large Language Models (LLMs) only by whether they select the correct answers. This binary scorin
arXiv:2608.02919v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in complex, compound AI systems where performance hinges on the quality of prompts. Recent state-