Parameter-efficient fine-tuning (PEFT) has become a popular way to adapt large pre-trained models to new tasks. Most PEFT methods update only a small subset of parameters while freezing the rest, avoiding redundant computation. As they maximize the absolute size of the updates without regard to the parameters' original scale, the resulting changes in model behavior can be minimal. In contrast, we maximize updates relative to each parameter's scale, yielding more meaningful downstream adaptation. We propose Gradient-to-Weight Ratio and Entropy-guided Masking (GEM), a parameter scale-aware, distribution-sensitive sparse fine-tuning framework. GEM prioritizes parameters whose updates are significant in proportion to their initial pre-trained values. It also adaptively determines how many parameters to tune at each layer based on the entropy of parameter values, thereby making the most effective use of the computational budget in PEFT. Our empirical study demonstrates the efficacy of GEM on both general-domain tasks (GLUE and SuperGLUE) and domain-specific tasks (GSM8k and MBPP), achieving up to a 1.6% improvement in fine-tuning accuracy over full fine-tuning while updating only 0.1% of model parameters.

Comics offer a compelling yet under-explored domain for computational narrative analysis, combining text and imagery in ways distinct from purely textual or audiovisual media. We introduce ComicScene154, a manually annotated dataset of scene-level narrative arcs derived from public-domain comic books spanning diverse genres. By conceptualizing comics as an abstraction for narrative-driven, multimodal data, we highlight their potential to inform broader research on multi-modal storytelling. To demonstrate the utility of ComicScene154, we present a baseline scene segmentation pipeline, providing an initial benchmark that future studies can build upon. Our results indicate that ComicScene154 constitutes a valuable resource for advancing computational methods in multimodal narrative understanding and expanding the scope of comic analysis within the Natural Language Processing community.

Cross-organizational collaboration in Model-Based Systems Engineering (MBSE) faces many challenges in achieving semantic alignment across independently developed system models. SysML v2 introduces enhanced structural modularity and formal semantics, offering a stronger foundation for interoperable modeling. Meanwhile, GPT-based Large Language Models (LLMs) provide new capabilities for assisting model understanding and integration. This paper proposes a structured, prompt-driven approach for LLM-assisted semantic alignment of SysML v2 models. The core contribution lies in the iterative development of an alignment approach and interaction prompts, incorporating model extraction, semantic matching, and verification. The approach leverages SysML v2 constructs such as alias, import, and metadata extensions to support traceable, soft alignment integration. It is demonstrated with a GPT-based LLM through an example of a measurement system. Benefits and limitations are discussed.

Usability describes a set of essential quality attributes of user interfaces (UI) that influence human-computer interaction. Common evaluation methods, such as usability testing and inspection, are effective but resource-intensive and require expert involvement. This makes them less accessible for smaller organizations. Recent advances in multimodal LLMs offer promising opportunities to automate usability evaluation processes partly by analyzing textual, visual, and structural aspects of software interfaces. To investigate this possibility, we formulate usability evaluation as a recommendation task, where multimodal LLMs rank usability issues by severity. We conducted an initial proof-of-concept study to compare LLM-generated usability improvement recommendations with usability expert assessments. Our findings indicate the potential of LLMs to enable faster and more cost-effective usability evaluation, which makes it a practical alternative in contexts with limited expert resources.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable multimodal understanding capabilities in Visual Question Answering (VQA) tasks by integrating visual and textual features. However, under the challenging ten-choice question evaluation paradigm, existing methods still exhibit significant limitations when processing PDF documents with complex layouts and lengthy content. Notably, current mainstream models suffer from a strong bias toward English training data, resulting in suboptimal performance for Japanese and other language scenarios. To address these challenges, this paper proposes a novel Japanese PDF document understanding framework that combines multimodal hierarchical reasoning mechanisms with Colqwen-optimized retrieval methods, while innovatively introducing a semantic verification strategy through sub-question decomposition. Experimental results demonstrate that our framework not only significantly enhances the model's deep semantic parsing capability for complex documents, but also exhibits superior robustness in practical application scenarios.

-- Daily life support robots must interpret ambiguous verbal instructions involving demonstratives such as "Bring me that cup," even when objects or users are out of the robot's view. Existing approaches to exophora resolution primarily rely on visual data and thus fail in real-world scenarios where the object or user is not visible. We propose Multimodal Interactive Exophora resolution with user Localization (MIEL), which is a multimodal exophora resolution framework leveraging sound source localization (SSL), semantic mapping, visual-language models (VLMs), and interactive questioning with GPT -4o. SSL is utilized to orient the robot toward users who are initially outside its visual field, enabling accurate identification of user gestures and pointing directions. When ambiguities remain, the robot proactively interacts with the user, employing GPT -4o to formulate clarifying questions. Experiments in a real-world environment showed results that were approximately 1.3 times better when the user was visible to the robot and 2.0 times better when the user was not visible to the robot, compared to the methods without SSL and interactive questioning. In our daily life, we frequently use verbal instructions that include demonstratives, such as "Take that for me," but for robots, the target object is often unclear and the user or object is often not in the robot's view. One of the challenges in the field of robotics is enabling daily life support robots to understand and execute tasks based on such instructions and situations [1]. To achieve this, implementing exophora resolution [2], [3] is essential. Exophora resolution involves identifying the referent --whether a person or object -- associated with anaphora (demonstratives or pronouns) within utterances, based on the surrounding context of the speaker or listener. For instance, if a user instructs the robot to "Bring me that cup," the robot must identify the target object corresponding to "that cup," even if there are many cups in the environment.

Large Language Models (LLMs) have shown significant progress in Open-domain question answering (ODQA), yet most evaluations focus on English and assume locale-invariant answers across languages. This assumption neglects the cultural and regional variations that affect question understanding and answer, leading to biased evaluation in multilingual benchmarks. To address these limitations, we introduce XLQA, a novel benchmark explicitly designed for locale-sensitive multilingual ODQA. XLQA contains 3,000 English seed questions expanded to eight languages, with careful filtering for semantic consistency and human-verified annotations distinguishing locale-invariant and locale-sensitive cases. Our evaluation of five state-of-the-art multilingual LLMs reveals notable failures on locale-sensitive questions, exposing gaps between English and other languages due to a lack of locale-grounding knowledge. We provide a systematic framework and scalable methodology for assessing multilingual QA under diverse cultural contexts, offering a critical resource to advance the real-world applicability of multilingual ODQA systems. Our findings suggest that disparities in training data distribution contribute to differences in both linguistic competence and locale-awareness across models.

Large Language Models (LLMs) confront significant memory challenges due to the escalating KV cache with increasing sequence length. As a crucial technique, existing cross-layer KV cache sharing methods either necessitate modified model architectures with subsequent pre-training or incur significant performance degradation at high compression rates. To mitigate these challenges, we propose CommonKV, a training-free method for cross-layer KV cache compression through adjacent parameters sharing. Inspired by the high similarity observed in cross-layer hidden states, we utilize Singular Value Decomposition (SVD) to achieve weight sharing across adjacent parameters, resulting in a more easily mergeable latent KV cache. Furthermore, we also introduce an adaptive budget allocation strategy. It dynamically assigns compression budgets based on cosine similarity, ensuring that dissimilar caches are not over-compressed. Experiments across multiple backbone models and benchmarks including LongBench and Ruler demonstrate that the proposed method consistently outperforms existing low-rank and cross-layer approaches at various compression ratios. Moreover, we find that the benefits of CommonKV are orthogonal to other quantization and eviction methods. By integrating these approaches, we can ultimately achieve a 98\% compression ratio without significant performance loss.

Code completion entails the task of providing missing tokens given a surrounding context. It can boost developer productivity while providing a powerful code discovery tool. Following the Large Language Model (LLM) wave, code completion has been approached with diverse LLMs fine-tuned on code (code LLMs). The performance of code LLMs can be assessed with downstream and intrinsic metrics. Downstream metrics are usually employed to evaluate the practical utility of a model, but can be unreliable and require complex calculations and domain-specific knowledge. In contrast, intrinsic metrics such as perplexity, entropy, and mutual information, which measure model confidence or uncertainty, are simple, versatile, and universal across LLMs and tasks, and can serve as proxies for functional correctness and hallucination risk in LLM-generated code. Motivated by this, we evaluate the confidence of LLMs when generating code by measuring code perplexity across programming languages, models, and datasets using various LLMs, and a sample of 1008 files from 657 GitHub projects. We find that strongly-typed languages exhibit lower perplexity than dynamically typed languages. Scripting languages also demonstrate higher perplexity. Perl appears universally high in perplexity, whereas Java appears low. Code perplexity depends on the employed LLM, but not on the code dataset. Although code comments often increase perplexity, the language ranking based on perplexity is barely affected by their presence. LLM researchers, developers, and users can employ our findings to assess the benefits and suitability of LLM-based code completion in specific software projects based on how language, model choice, and code characteristics impact model confidence.

Building upon the strong sequence modeling capability, Generative Recommendation (GR) has gradually assumed a dominant position in the application of recommendation tasks (e.g., video and product recommendation). However, the application of Generative Recommendation in Point-of-Interest (POI) recommendation, where user preferences are significantly affected by spatiotemporal variations, remains a challenging open problem. In this paper, we propose Spacetime-GR, the first spacetime-aware generative model for large-scale online POI recommendation. It extends the strong sequence modeling ability of generative models by incorporating flexible spatiotemporal information encoding. Specifically, we first introduce a geographic-aware hierarchical POI indexing strategy to address the challenge of large vocabulary modeling. Subsequently, a novel spatiotemporal encoding module is introduced to seamlessly incorporate spatiotemporal context into user action sequences, thereby enhancing the model's sensitivity to spatiotemporal variations. Furthermore, we incorporate multimodal POI embeddings to enrich the semantic understanding of each POI. Finally, to facilitate practical deployment, we develop a set of post-training adaptation strategies after sufficient pre-training on action sequences. These strategies enable Spacetime-GR to generate outputs in multiple formats (i.e., embeddings, ranking scores and POI candidates) and support a wide range of downstream application scenarios (i.e., ranking and end-to-end recommendation). We evaluate the proposed model on both public benchmark datasets and large-scale industrial datasets, demonstrating its superior performance over existing methods in terms of POI recommendation accuracy and ranking quality. Furthermore, the model is the first generative model deployed in online POI recommendation services that scale to hundreds of millions of POIs and users.