Knowledge catalogue
arXiv:2606.10445v1 Announce Type: cross Abstract: Semi-structured 2:4 sparsity is widely supported by modern accelerators, providing up to a 2x theoretical speedup. However, its strict 50% sparsity co
arXiv:2406.08726v3 Announce Type: replace Abstract: Large language models (LLMs) generate text that reinforces standard language ideology: a bias towards certain language varieties that are granted mo
arXiv:2606.09877v1 Announce Type: cross Abstract: LLM wiki systems compile knowledge into pre-filled KV caches for efficient inference, but assume a static corpus -- an assumption that fails whenever
arXiv:2512.02201v3 Announce Type: replace Abstract: This paper introduces Swivuriso, a 3000-hour multilingual speech dataset developed as part of the African Next Voices project, to support the develo
arXiv:2606.10316v1 Announce Type: new Abstract: Spreadsheets and tables are widely used representations for structured data analysis, but effective analysis still requires substantial manual effort an
arXiv:2606.10327v1 Announce Type: new Abstract: Automated Essay Scoring (AES) systems must judge interdependent discourse elements (e.g., lead, claim, evidence, conclusion), yet most approaches treat
arXiv:2606.11082v1 Announce Type: new Abstract: This study investigates cross-lingual distributional skew (the Shibboleth Effect) in frontier large language models (LLMs) subjected to sustained advers
arXiv:2606.10921v1 Announce Type: new Abstract: Long-document question answering (QA) requires large language models (LLMs) to reason over evidence scattered across lengthy documents, where answers of
arXiv:2606.10860v1 Announce Type: cross Abstract: Production LLMs receive instructions from sources with very different levels of trust, yet attend to every token with uniform architectural privilege.
arXiv:2606.10520v1 Announce Type: new Abstract: Post-training quantization at the 2-bit level enables low-cost deployment and inference acceleration for large language models (LLMs). Scalar quantizati
arXiv:2606.09570v2 Announce Type: replace Abstract: As AI assistants serve millions of users daily, evaluating user experience (UX) beyond general model capability has become increasingly important. W
arXiv:2606.11079v1 Announce Type: new Abstract: Evaluation remains a critical bottleneck for interactive agent development. Existing evaluation methods often rely on static benchmarks, which fail to c
arXiv:2606.10423v1 Announce Type: new Abstract: Autonomous web navigation remains challenging for LLM agents, and the strongest generalist systems rely on proprietary reasoning models whose inference
arXiv:2501.14717v2 Announce Type: replace Abstract: Table modeling has progressed for decades. In this work, we revisit this trajectory and highlight emerging challenges in the LLM era, particularly t
arXiv:2606.09421v2 Announce Type: replace Abstract: Large language model agents increasingly rely on skills: reusable procedural documents encoding workflows, tool use, implementation patterns, valida
arXiv:2606.10287v1 Announce Type: cross Abstract: Evaluating Knowledge Graph Completion (KGC) models remains challenging because standard assessment relies on isolated rank-based metrics such as MRR,
arXiv:2606.10302v1 Announce Type: new Abstract: Open-ended generation tasks often require a set of meaningfully different outputs, yet large language models often produce similar generations. Existing
arXiv:2606.10428v1 Announce Type: new Abstract: We investigate whether commonly available LoRA variants have an advantage over basic LoRA in multilingual instruction tuning. Experiments involving LoRA
arXiv:2606.11009v1 Announce Type: new Abstract: Large language models are increasingly used to adapt math word problems for personalized learning at scale, but it remains an open question whether thos
arXiv:2603.28054v2 Announce Type: replace Abstract: In this paper, we introduce GhostWriteBench, a dataset for LLM authorship attribution. It comprises long-form texts (50K+ words per book) generated
arXiv:2603.06915v2 Announce Type: replace Abstract: The extraction of structured information from raw text is a fundamental component of many NLP applications, including document retrieval, ranking, a
arXiv:2606.06758v1 Announce Type: new Abstract: Final-answer accuracy, retrieval recall, and citation overlap do not by themselves identify whether a long-context or retrieval-augmented language model
arXiv:2601.08097v2 Announce Type: replace Abstract: Reward modeling is essential for aligning large language models with human preferences, yet predominant architectures rely on a static pooling strat
arXiv:2606.07219v1 Announce Type: new Abstract: The convergence of large language models and social bots allows malicious actors to manipulate the information ecosystem by generating human-like conten
arXiv:2606.07513v1 Announce Type: new Abstract: Humans learn from social life. Simulating this process with LLM-powered agents represents a promising research direction, raising a natural question: wh
arXiv:2606.06879v1 Announce Type: new Abstract: Our prior work introduced COVA, a synthetically generated multi-turn conversational smishing dataset of 3,201 labeled conversations, establishing baseli
arXiv:2606.06865v1 Announce Type: new Abstract: Large language models (LLMs) have been increasingly explored for graph computation, where tasks require reasoning over structured relationships and algo
arXiv:2512.13278v2 Announce Type: replace Abstract: Agentic reinforcement learning has advanced large language models (LLMs) to reason through long chain-of-thought trajectories while interleaving ext
arXiv:2606.07040v1 Announce Type: new Abstract: Open-ended reward modeling requires judges that can follow subtle, domain-specific preferences when verifiable answers are unavailable. Existing rubric-
arXiv:2606.06985v1 Announce Type: new Abstract: Code-switching (CS), the alternation between multiple languages within a single utterance, remains challenging for Automatic Speech Recognition (ASR). T
arXiv:2606.06842v1 Announce Type: new Abstract: Table reasoning remains challenging for large language models (LLMs), particularly in tasks that require multi-step inference over long and structured t
arXiv:2512.09634v2 Announce Type: replace Abstract: This article presents the creation of an Estonian-language dataset for document-level subjectivity, analyzes the resulting annotations, and reports
arXiv:2601.10896v2 Announce Type: replace Abstract: LLMs are increasingly used as third-party judges, yet their reliability when evaluating speakers in dialogue remains poorly understood. We show that
arXiv:2606.07356v1 Announce Type: cross Abstract: Text-guided audio editing aims to modify the language-specified acoustic content while preserving edit-irrelevant source components. Existing training
arXiv:2606.06788v1 Announce Type: new Abstract: Evaluations of large language models (LLMs) in scientific information seeking tasks have become increasingly use-centric, such as conducting live or mul
arXiv:2606.07130v1 Announce Type: new Abstract: As AI systems become more widely adopted, the demand for factual and faithful generation grows. Properly attributing information through citations becom
arXiv:2606.07190v1 Announce Type: new Abstract: Reasoning prefixes shape the future trajectory of LLM problem solving, yet existing process reward models usually evaluate them through local step corre
arXiv:2606.07183v1 Announce Type: new Abstract: This work examines the semantic geometry underlying NLP models. We compare supervised vector embeddings, such as CamemBERT, with lexical co-occurrence g
arXiv:2606.07218v1 Announce Type: cross Abstract: Multi-hop RAG poses a data-engineering problem beyond passage matching: under fixed retrieval budgets, a system must organize retrieved text into evid
arXiv:2606.06586v1 Announce Type: new Abstract: Large language models (LLMs) trained predominantly on English data encode substantial world knowledge, yet often fail to express it reliably in other la
arXiv:2606.06857v1 Announce Type: new Abstract: A central goal of cognitive neuroscience is to characterize the features that are represented by human language cortex. Artificial language models (LMs)
arXiv:2606.07240v1 Announce Type: new Abstract: Cross-lingual voice cloning aims to generate speech in a target language while preserving speaker identity from a source-language reference. This task i
arXiv:2606.06797v1 Announce Type: new Abstract: Cultural-aspect work on large language models is dominated by a negative target: which outputs to suppress. We argue that a constructive counterpart is
arXiv:2606.07123v1 Announce Type: new Abstract: Social meaning in language is inherently perspectival, varying across annotator backgrounds, demographics, and ideological positions. However, most NLP
arXiv:2606.07342v1 Announce Type: new Abstract: Adapting large language models (LLMs) to clinical workflows often requires costly fine-tuning or manual prompt and pipeline engineering. We study LLM-gu
arXiv:2606.07402v1 Announce Type: new Abstract: Language agents are increasingly deployed over accumulating multimodal information, yet existing benchmarks assume a human-human form with sparse visual
arXiv:2606.07020v1 Announce Type: new Abstract: Multilingual and multicultural benchmarks now cover dozens of languages and model families, but the resulting score landscapes remain metric-rich and in
arXiv:2606.06754v1 Announce Type: cross Abstract: We present MADRAG, a training-free framework for analytic essay scoring that combines multi-agent reasoning with retrieval-augmented grounding. Unlike
arXiv:2602.14209v2 Announce Type: replace-cross Abstract: Block diffusion LLMs are an emerging paradigm for parallel language generation, but their KV caching makes memory access the dominant bottlene
arXiv:2606.07057v1 Announce Type: cross Abstract: Evaluating the quality of automatically generated keyphrases remains a complex challenge. Traditional metrics either rely on exact lexical matching or
arXiv:2602.11201v2 Announce Type: replace Abstract: Chain-of-Thought (CoT) explanations are widely used to interpret how language models solve complex problems, yet it remains unclear whether these st
arXiv:2511.07380v2 Announce Type: replace Abstract: Adapting large language models (LLMs) to low-resource domains remains challenging due to the scarcity of domain-specific data. While in-domain data
arXiv:2606.07229v1 Announce Type: cross Abstract: We introduce MMAE, a Massive Multitask Audio Editing benchmark, serving as the first comprehensive evaluation testbed designed for general-purpose ins
arXiv:2606.07069v1 Announce Type: new Abstract: We introduce mmPISA-bench, a compact high-quality multilingual reasoning benchmark derived from the OECD Programme for International Student Assessment
arXiv:2606.07066v1 Announce Type: new Abstract: Semantic association between a word and its context has been identified as an important component of reading comprehension, even when word predictabilit
arXiv:2606.06738v1 Announce Type: new Abstract: Building monolingual language models (LMs) for low-resource languages typically relies on adapting pretrained language models (PLMs) by finetuning the w
arXiv:2606.06573v1 Announce Type: cross Abstract: We introduce scale-selective Proper Orthogonal Decomposition (POD) for transformer attention fields, inspired by the use of POD for extracting energet
arXiv:2606.07300v1 Announce Type: new Abstract: Language is a vehicle for thought, intricately tied to sounds, symbols, and meaning. However, most large language model (LLM) research focuses on meanin
arXiv:2502.00527v2 Announce Type: replace-cross Abstract: The KV cache in large language models is a dominant factor in memory usage, limiting their broader applicability. Quantizing the cache to lowe
arXiv:2606.06994v1 Announce Type: new Abstract: What makes a sentence encoder produce good concept representations? We approach this through the lens of representational compositionality: an encoder s