Large Language Models (LLMs) have demonstrated remarkable capabilities across numerous tasks, yet principled explanations for their underlying mechanisms and several phenomena, such as scaling laws, hallucinations, and related behaviors, remain elusive. In this work, we revisit the classical relationship between compression and prediction, grounded in Kolmogorov complexity and Shannon information theory, to provide deeper insights into LLM behaviors. By leveraging the Kolmogorov Structure Function and interpreting LLM compression as a two-part coding process, we offer a detailed view of how LLMs acquire and store information across increasing model and data scales -- from pervasive syntactic patterns to progressively rarer knowledge elements. Motivated by this theoretical perspective and natural assumptions inspired by Heap's and Zipf's laws, we introduce a simplified yet representative hierarchical data-generation framework called the Syntax-Knowledge model. Under the Bayesian setting, we show that prediction and compression within this model naturally lead to diverse learning and scaling behaviors observed in LLMs. In particular, our theoretical analysis offers intuitive and principled explanations for both data and model scaling laws, the dynamics of knowledge acquisition during training and fine-tuning, factual knowledge hallucinations in LLMs. The experimental results validate our theoretical predictions.

This paper presents Team Marikarp's solution for the SIGIR 2025 LiveRAG competition. The competition's evaluation set, automatically generated by DataMorgana from internet corpora, encompassed a wide range of target topics, question types, question formulations, audience types, and knowledge organization methods. It offered a fair evaluation of retrieving question-relevant supporting documents from a 15M documents subset of the FineWeb corpus. Our proposed knowledge-aware diverse reranking RAG pipeline achieved first place in the competition.

The performance of Large Language Models (LLMs) is intrinsically linked to the quality of its training data. Although several studies have proposed methods for high-quality data selection, they do not consider the importance of knowledge richness in text corpora. In this paper, we propose a novel and gradient-free High-Knowledge Scorer (HKS) to select high-quality data from the dimension of knowledge, to alleviate the problem of knowledge scarcity in the pre-trained corpus. We propose a comprehensive multi-domain knowledge element pool and introduce knowledge density and coverage as metrics to assess the knowledge content of the text. Based on this, we propose a comprehensive knowledge scorer to select data with intensive knowledge, which can also be utilized for domain-specific high-knowledge data selection by restricting knowledge elements to the specific domain. We train models on a high-knowledge bilingual dataset, and experimental results demonstrate that our scorer improves the model's performance in knowledge-intensive and general comprehension tasks, and is effective in enhancing both the generic and domain-specific capabilities of the model.

Machine learning (ML) and artificial intelligence (AI) conferences including NeurIPS and ICML have experienced a significant decline in peer review quality in recent years. To address this growing challenge, we introduce the Isotonic Mechanism, a computationally efficient approach to enhancing the accuracy of noisy review scores by incorporating authors' private assessments of their submissions. Under this mechanism, authors with multiple submissions are required to rank their papers in descending order of perceived quality. Subsequently, the raw review scores are calibrated based on this ranking to produce adjusted scores. We prove that authors are incentivized to truthfully report their rankings because doing so maximizes their expected utility, modeled as an additive convex function over the adjusted scores. Moreover, the adjusted scores are shown to be more accurate than the raw scores, with improvements being particularly significant when the noise level is high and the author has many submissions -- a scenario increasingly prevalent at large-scale ML/AI conferences. We further investigate whether submission quality information beyond a simple ranking can be truthfully elicited from authors. We establish that a necessary condition for truthful elicitation is that the mechanism be based on pairwise comparisons of the author's submissions. This result underscores the optimality of the Isotonic Mechanism, as it elicits the most fine-grained truthful information among all mechanisms we consider. We then present several extensions, including a demonstration that the mechanism maintains truthfulness even when authors have only partial rather than complete information about their submission quality. Finally, we discuss future research directions, focusing on the practical implementation of the mechanism and the further development of a theoretical framework inspired by our mechanism.

In human-AI interactions, explanation is widely seen as necessary for enabling trust in AI systems. We argue that trust, however, may be a pre-requisite because explanation is sometimes impossible. We derive this result from a formalization of explanation as a search process through knowledge networks, where explainers must find paths between shared concepts and the concept to be explained, within finite time. Our model reveals that explanation can fail even under theoretically ideal conditions - when actors are rational, honest, motivated, can communicate perfectly, and possess overlapping knowledge. This is because successful explanation requires not just the existence of shared knowledge but also finding the connection path within time constraints, and it can therefore be rational to cease attempts at explanation before the shared knowledge is discovered. This result has important implications for human-AI interaction: as AI systems, particularly Large Language Models, become more sophisticated and able to generate superficially compelling but spurious explanations, humans may default to trust rather than demand genuine explanations. This creates risks of both misplaced trust and imperfect knowledge integration.

Novelty evaluation is vital for the promotion and management of innovation. With the advancement of information techniques and the open data movement, some progress has been made in novelty measurements. Tracking and reviewing novelty measures provides a data-driven way to assess contributions, progress, and emerging directions in the science field. As academic papers serve as the primary medium for the dissemination, validation, and discussion of scientific knowledge, this review aims to offer a systematic analysis of novelty measurements for scientific papers. We began by comparing the differences between scientific novelty and four similar concepts, including originality, scientific innovation, creativity, and scientific breakthrough. Next, we reviewed the types of scientific novelty. Then, we classified existing novelty measures according to data types and reviewed the measures for each type. Subsequently, we surveyed the approaches employed in validating novelty measures and examined the current tools and datasets associated with these measures. Finally, we proposed several open issues for future studies.

Conversational grounding is a collaborative mechanism for establishing mutual knowledge among participants engaged in a dialogue. This experimental study analyzes information-seeking conversations to investigate the capabilities of large language models in classifying dialogue turns related to explicit or implicit grounding and predicting grounded knowledge elements. Our experimental results reveal challenges encountered by large language models in the two tasks and discuss ongoing research efforts to enhance large language model-based conversational grounding comprehension through pipeline architectures and knowledge bases. These initiatives aim to develop more effective dialogue systems that are better equipped to handle the intricacies of grounded knowledge in conversations.

We present Visual Knowledge oriented Programming platform (VisKoP), a knowledge base question answering (KBQA) system that integrates human into the loop to edit and debug the knowledge base (KB) queries. VisKoP not only provides a neural program induction module, which converts natural language questions into knowledge oriented program language (KoPL), but also maps KoPL programs into graphical elements. KoPL programs can be edited with simple graphical operators, such as dragging to add knowledge operators and slot filling to designate operator arguments. Moreover, VisKoP provides auto-completion for its knowledge base schema and users can easily debug the KoPL program by checking its intermediate results. To facilitate the practical KBQA on a million-entity-level KB, we design a highly efficient KoPL execution engine for the back-end. Experiment results show that VisKoP is highly efficient and user interaction can fix a large portion of wrong KoPL programs to acquire the correct answer. The VisKoP online demo https://demoviskop.xlore.cn (Stable release of this paper) and https://viskop.xlore.cn (Beta release with new features), highly efficient KoPL engine https://pypi.org/project/kopl-engine, and screencast video https://youtu.be/zAbJtxFPTXo are now publicly available.

Computing competency is becoming an essential quality needed by industry. For decades, the gap between baccalaureate computing graduates and industry needs was a discussion topic. Most graduates seek employment in deference to continuing their full-time graduate (master's or doctoral) programs. While the percent of such choice varies by institution, it is estimated that about 5% of computing graduates choose full-time graduate study upon graduation, meaning that 95% of computing graduates seek jobs in business, government, or industry.15 While computing graduates may acquire jobs in today's world, they often lack the competencies (skills and dispositions) expected in the workplace. Most undergraduate computing-degree programs want to produce job-ready graduates who are productive on the first workday. They often seek local advisory boards composed of industry, government, and business representatives to help develop a functional computing curriculum for their students. Information technology and computing disciplines are changing, and new fields appear continuously. Computing curricula and undergraduate programs are challenged to keep up with this rapid change. Employers are looking for competent graduates who can apply the knowledge, skill, and culture they acquire in college to solve problems as soon as they enter the workforce. High-performance computing (HPC) and parallel and distributed computing (PDC) have become pervasive.

Knowledge (including structured knowledge such as schema and ontology, and unstructured knowledge such as web corpus) is a critical part of dialog understanding, especially for unseen tasks and domains. Traditionally, such domain-specific knowledge is encoded implicitly into model parameters for the execution of downstream tasks, which makes training inefficient. In addition, such models are not easily transferable to new tasks with different schemas. In this work, we propose to perform dialog state tracking grounded on knowledge encoded externally. We query relevant knowledge of various forms based on the dialog context where such information can ground the prediction of dialog states. We demonstrate superior performance of our proposed method over strong baselines, especially in the few-shot learning setting.