In spite of the remarkable advancements in the field of Natural Language Processing, the task of Entity Linking (EL) remains challenging in the field of humanities due to complex document typologies, lack of domain-specific datasets and models, and long-tail entities, i.e., entities under-represented in Knowledge Bases (KBs). The goal of this paper is to address these issues with two main contributions. The first contribution is DELICATE, a novel neuro-symbolic method for EL on historical Italian which combines a BERT-based encoder with contextual information from Wikidata to select appropriate KB entities using temporal plausibility and entity type consistency. The second contribution is ENEIDE, a multi-domain EL corpus in historical Italian semi-automatically extracted from two annotated editions spanning from the 19th to the 20th century and including literary and political texts. Results show how DELICATE outperforms other EL models in historical Italian even if compared with larger architectures with billions of parameters. Moreover, further analyses reveal how DELICATE confidence scores and features sensitivity provide results which are more explainable and interpretable than purely neural methods.

Autonomous agents powered by large language models (LLMs) have the potential to significantly enhance human productivity by reasoning, using tools, and executing complex tasks in diverse environments. However, current approaches to developing such agents remain costly and inefficient, as they typically require manually constructed task datasets and reinforcement learning (RL) pipelines with extensive random exploration. These limitations lead to prohibitively high data-construction costs, low exploration efficiency, and poor sample utilization. To address these challenges, we present AgentEvolver, a self-evolving agent system that leverages the semantic understanding and reasoning capabilities of LLMs to drive autonomous agent learning. AgentEvolver introduces three synergistic mechanisms: (i) self-questioning, which enables curiosity-driven task generation in novel environments, reducing dependence on handcrafted datasets; (ii) self-navigating, which improves exploration efficiency through experience reuse and hybrid policy guidance; and (iii) self-attributing, which enhances sample efficiency by assigning differentiated rewards to trajectory states and actions based on their contribution. By integrating these mechanisms into a unified framework, AgentEvolver enables scalable, cost-effective, and continual improvement of agent capabilities. Preliminary experiments indicate that AgentEvolver achieves more efficient exploration, better sample utilization, and faster adaptation compared to traditional RL-based baselines.

Existing work investigates the reasoning capabilities of large language models (LLMs) to uncover their limitations, human-like biases and underlying processes. Such studies include evaluations of base LLMs (pre-trained on unlabeled corpora only) for this purpose. Our position paper argues that evaluating base LLMs' reasoning capabilities raises inherent methodological concerns that are overlooked in such existing studies. We highlight the fundamental mismatch between base LLMs' pretraining objective and normative qualities, such as correctness, by which reasoning is assessed. In particular, we show how base LLMs generate logically valid or invalid conclusions as coincidental byproducts of conforming to purely linguistic patterns of statistical plausibility. This fundamental mismatch challenges the assumptions that (a) base LLMs' outputs can be assessed as their bona fide attempts at correct answers or conclusions; and (b) conclusions about base LLMs' reasoning can generalize to post-trained LLMs optimized for successful instruction-following. We call for a critical re-examination of existing work that relies implicitly on these assumptions, and for future work to account for these methodological pitfalls.

Retrieval-Augmented Generation (RAG) has emerged as a powerful framework for enhancing the capabilities of Large Language Models (LLMs) by integrating retrieval-based methods with generative models. As external knowledge repositories continue to expand and the parametric knowledge within models becomes outdated, a critical challenge for RAG systems is resolving conflicts between retrieved external information and LLMs' internal knowledge, which can significantly compromise the accuracy and reliability of generated content. However, existing approaches to conflict resolution typically operate at the token or semantic level, often leading to fragmented and partial understanding of factual discrepancies between LLMs' knowledge and context, particularly in knowledge-intensive tasks. To address this limitation, we propose TruthfulRAG, the first framework that leverages Knowledge Graphs (KGs) to resolve factual-level knowledge conflicts in RAG systems. Specifically, TruthfulRAG constructs KGs by systematically extracting triples from retrieved content, utilizes query-based graph retrieval to identify relevant knowledge, and employs entropy-based filtering mechanisms to precisely locate conflicting elements and mitigate factual inconsistencies, thereby enabling LLMs to generate faithful and accurate responses. Extensive experiments reveal that TruthfulRAG outperforms existing methods, effectively alleviating knowledge conflicts and improving the robustness and trustworthiness of RAG systems.

Cultural Heritage texts contain rich knowledge that is difficult to query systematically due to the challenges of converting unstructured discourse into structured Knowledge Graphs (KGs). This paper introduces ATR4CH (Adaptive Text-to-RDF for Cultural Heritage), a systematic five-step methodology for Large Language Model-based Knowledge Extraction from Cultural Heritage documents. We validate the methodology through a case study on authenticity assessment debates. Methodology - ATR4CH combines annotation models, ontological frameworks, and LLM-based extraction through iterative development: foundational analysis, annotation schema development, pipeline architecture, integration refinement, and comprehensive evaluation. We demonstrate the approach using Wikipedia articles about disputed items (documents, artifacts...), implementing a sequential pipeline with three LLMs (Claude Sonnet 3.7, Llama 3.3 70B, GPT-4o-mini). Findings - The methodology successfully extracts complex Cultural Heritage knowledge: 0.96-0.99 F1 for metadata extraction, 0.7-0.8 F1 for entity recognition, 0.65-0.75 F1 for hypothesis extraction, 0.95-0.97 for evidence extraction, and 0.62 G-EVAL for discourse representation. Smaller models performed competitively, enabling cost-effective deployment. Originality - This is the first systematic methodology for coordinating LLM-based extraction with Cultural Heritage ontologies. ATR4CH provides a replicable framework adaptable across CH domains and institutional resources. Research Limitations - The produced KG is limited to Wikipedia articles. While the results are encouraging, human oversight is necessary during post-processing. Practical Implications - ATR4CH enables Cultural Heritage institutions to systematically convert textual knowledge into queryable KGs, supporting automated metadata enrichment and knowledge discovery.

To improve Multi-step Mathematical Reasoning (MsMR) of Large Language Models (LLMs), it is crucial to obtain scalable supervision from the corpus by automatically critiquing mistakes in the reasoning process of MsMR and rendering a final verdict of the problem-solution. Most existing methods rely on crafting high-quality supervised fine-tuning demonstrations for critiquing capability enhancement and pay little attention to delving into the underlying reason for the poor critiquing performance of LLMs. In this paper, we orthogonally quantify and investigate the potential reason -- imbalanced evaluation preference, and conduct a statistical preference analysis. Motivated by the analysis of the reason, a novel perplexity-aware reinforcement learning algorithm is proposed to rectify the evaluation preference, elevating the critiquing capability. Specifically, to probe into LLMs' critiquing characteristics, a One-to-many Problem-Solution (OPS) benchmark is meticulously constructed to quantify the behavior difference of LLMs when evaluating the problem solutions generated by itself and others. Then, to investigate the behavior difference in depth, we conduct a statistical preference analysis oriented on perplexity and find an intriguing phenomenon -- ``LLMs incline to judge solutions with lower perplexity as correct'', which is dubbed as \textit{imbalanced evaluation preference}. To rectify this preference, we regard perplexity as the baton in the algorithm of Group Relative Policy Optimization, supporting the LLMs to explore trajectories that judge lower perplexity as wrong and higher perplexity as correct. Extensive experimental results on our built OPS and existing available critic benchmarks demonstrate the validity of our method.

Despite the remarkable success of the LLaVA architecture for vision-language tasks, its design inherently struggles to effectively integrate visual features due to the inherent mismatch between text and vision modalities. We tackle this issue from a novel perspective in which the LLM not only serves as a language model but also a powerful vision encoder. To this end, we present LLaViT - Large Language Models as extended Vision Transformers - which enables the LLM to simultaneously function as a vision encoder through three key modifications: (1) learning separate QKV projections for vision modality, (2) enabling bidirectional attention on visual tokens, and (3) incorporating both global and local visual representations. Through extensive controlled experiments on a wide range of LLMs, we demonstrate that LLaViT significantly outperforms the baseline LLaVA method on a multitude of benchmarks, even surpassing models with double its parameter count, establishing a more effective approach to vision-language modeling.

We introduce Music Flamingo, a novel large audio-language model designed to advance music (including song) understanding in foundational audio models. While audio-language research has progressed rapidly, music remains challenging due to its dynamic, layered, and information-dense nature. Progress has been further limited by the difficulty of scaling open audio understanding models, primarily because of the scarcity of high-quality music data and annotations. As a result, prior models are restricted to producing short, high-level captions, answering only surface-level questions, and showing limited generalization across diverse musical cultures. To address these challenges, we curate MF-Skills, a large-scale dataset labeled through a multi-stage pipeline that yields rich captions and question-answer pairs covering harmony, structure, timbre, lyrics, and cultural context. We fine-tune an enhanced Audio Flamingo 3 backbone on MF-Skills and further strengthen multiple skills relevant to music understanding. To improve the model's reasoning abilities, we introduce a post-training recipe: we first cold-start with MF-Think, a novel chain-of-thought dataset grounded in music theory, followed by GRPO-based reinforcement learning with custom rewards. Music Flamingo achieves state-of-the-art results across 10+ benchmarks for music understanding and reasoning, establishing itself as a generalist and musically intelligent audio-language model. Beyond strong empirical results, Music Flamingo sets a new standard for advanced music understanding by demonstrating how models can move from surface-level recognition toward layered, human-like perception of songs. We believe this work provides both a benchmark and a foundation for the community to build the next generation of models that engage with music as meaningfully as humans do.

Fake news detection methods based on writing style have achieved remarkable progress. However, as adversaries increasingly imitate the style of authentic news, the effectiveness of such approaches is gradually diminishing. Recent research has explored incorporating large language models (LLMs) to enhance fake news detection. Yet, despite their transformative potential, LLMs remain an untapped goldmine for fake news detection, with their real-world adoption hampered by shallow functionality exploration, ambiguous usability, and prohibitive inference costs. In this paper, we propose a novel fake news detection framework, dubbed FactGuard, that leverages LLMs to extract event-centric content, thereby reducing the impact of writing style on detection performance. Furthermore, our approach introduces a dynamic usability mechanism that identifies contradictions and ambiguous cases in factual reasoning, adaptively incorporating LLM advice to improve decision reliability. To ensure efficiency and practical deployment, we employ knowledge distillation to derive FactGuard-D, enabling the framework to operate effectively in cold-start and resource-constrained scenarios. Comprehensive experiments on two benchmark datasets demonstrate that our approach consistently outperforms existing methods in both robustness and accuracy, effectively addressing the challenges of style sensitivity and LLM usability in fake news detection.

Large Language Models (LLMs) have demonstrated remarkable capabilities in generating human-like text, yet their applicability to dialogue systems in computer games remains limited. This limitation arises from their substantial hardware requirements, latency constraints, and the necessity to maintain clearly defined knowledge boundaries within a game setting. In this paper, we propose a modular NPC dialogue system that leverages Small Language Models (SLMs), fine-tuned to encode specific NPC personas and integrated with runtime-swappable memory modules. These memory modules preserve character-specific conversational context and world knowledge, enabling expressive interactions and long-term memory without retraining or model reloading during gameplay. We comprehensively evaluate our system using three open-source SLMs: DistilGPT-2, TinyLlama-1.1B-Chat, and Mistral-7B-Instruct, trained on synthetic persona-aligned data and benchmarked on consumer-grade hardware. While our approach is motivated by applications in gaming, its modular design and persona-driven memory architecture hold significant potential for broader adoption in domains requiring expressive, scalable, and memory-rich conversational agents, such as virtual assistants, customer support bots, or interactive educational systems.