Artificial Intelligence (AI) has increasingly influenced modern society, recently in particular through significant advancements in Large Language Models (LLMs). However, high computational and storage demands of LLMs still limit their deployment in resource-constrained environments. Knowledge distillation addresses this challenge by training a small student model from a larger teacher model. Previous research has introduced several distillation methods for both generating training data and for training the student model. Despite their relevance, the effects of state-of-the-art distillation methods on model performance and explainability have not been thoroughly investigated and compared. In this work, we enlarge the set of available methods by applying critique-revision prompting to distillation for data generation and by synthesizing existing methods for training. For these methods, we provide a systematic comparison based on the widely used Commonsense Question-Answering (CQA) dataset. While we measure performance via student model accuracy, we employ a human-grounded study to evaluate explainability. We contribute new distillation methods and their comparison in terms of both performance and explainability. This should further advance the distillation of small language models and, thus, contribute to broader applicability and faster diffusion of LLM technology.

In order for robots to be useful, they must perform practically relevant tasks in the real world, outside of the lab. While vision-language-action (VLA) models have demonstrated impressive results for end-to-end robot control, it remains an open question how far such models can generalize in the wild. We describe $π_{0.5}$, a new model based on $π_{0}$ that uses co-training on heterogeneous tasks to enable broad generalization. $π_{0.5}$\ uses data from multiple robots, high-level semantic prediction, web data, and other sources to enable broadly generalizable real-world robotic manipulation. Our system uses a combination of co-training and hybrid multi-modal examples that combine image observations, language commands, object detections, semantic subtask prediction, and low-level actions. Our experiments show that this kind of knowledge transfer is essential for effective generalization, and we demonstrate for the first time that an end-to-end learning-enabled robotic system can perform long-horizon and dexterous manipulation skills, such as cleaning a kitchen or bedroom, in entirely new homes.

State space models (SSMs) have emerged as an efficient alternative to Transformer models for language modeling, offering linear computational complexity and constant memory usage as context length increases. However, despite their efficiency in handling long contexts, recent studies have shown that SSMs, such as Mamba models, generally underperform compared to Transformers in long-context understanding tasks. To address this significant shortfall and achieve both efficient and accurate long-context understanding, we propose LongMamba, a training-free technique that significantly enhances the long-context capabilities of Mamba models. LongMamba builds on our discovery that the hidden channels in Mamba can be categorized into local and global channels based on their receptive field lengths, with global channels primarily responsible for long-context capability. These global channels can become the key bottleneck as the input context lengthens. Specifically, when input lengths largely exceed the training sequence length, global channels exhibit limitations in adaptively extend their receptive fields, leading to Mamba's poor long-context performance. The key idea of LongMamba is to mitigate the hidden state memory decay in these global channels by preventing the accumulation of unimportant tokens in their memory. This is achieved by first identifying critical tokens in the global channels and then applying token filtering to accumulate only those critical tokens. Through extensive benchmarking across synthetic and real-world long-context scenarios, LongMamba sets a new standard for Mamba's long-context performance, significantly extending its operational range without requiring additional training. Our code is available at https://github.com/GATECH-EIC/LongMamba.

Foundation models, trained on vast amounts of data using self-supervised techniques, have emerged as a promising frontier for advancing artificial intelligence (AI) applications in medicine. This study evaluates three different vision-language foundation models (RAD-DINO, CheXagent, and BiomedCLIP) on their ability to capture fine-grained imaging features for radiology tasks. The models were assessed across classification, segmentation, and regression tasks for pneumothorax and cardiomegaly on chest radiographs. Self-supervised RAD-DINO consistently excelled in segmentation tasks, while text-supervised CheXagent demonstrated superior classification performance. BiomedCLIP showed inconsistent performance across tasks. A custom segmentation model that integrates global and local features substantially improved performance for all foundation models, particularly for challenging pneumothorax segmentation. The findings highlight that pre-training methodology significantly influences model performance on specific downstream tasks. For fine-grained segmentation tasks, models trained without text supervision performed better, while text-supervised models offered advantages in classification and interpretability. These insights provide guidance for selecting foundation models based on specific clinical applications in radiology.

Determining the most effective Large Language Model for code smell detection presents a complex challenge. This study introduces a structured methodology and evaluation matrix to tackle this issue, leveraging a curated dataset of code samples consistently annotated with known smells. The dataset spans four prominent programming languages Java, Python, JavaScript, and C++; allowing for cross language comparison. We benchmark two state of the art LLMs, OpenAI GPT 4.0 and DeepSeek-V3, using precision, recall, and F1 score as evaluation metrics. Our analysis covers three levels of detail: overall performance, category level performance, and individual code smell type performance. Additionally, we explore cost effectiveness by comparing the token based detection approach of GPT 4.0 with the pattern-matching techniques employed by DeepSeek V3. The study also includes a cost analysis relative to traditional static analysis tools such as SonarQube. The findings offer valuable guidance for practitioners in selecting an efficient, cost effective solution for automated code smell detection

Reinforcement learning (RL) has become the core post-training technique for large language models (LLMs). RL for LLMs involves two stages: generation and training. The LLM first generates samples online, which are then used to derive rewards for training. The conventional view holds that the colocated architecture, where the two stages share resources via temporal multiplexing, outperforms the disaggregated architecture, in which dedicated resources are assigned to each stage. However, in real-world deployments, we observe that the colocated architecture suffers from resource coupling, where the two stages are constrained to use the same resources. This coupling compromises the scalability and cost-efficiency of colocated RL in large-scale training. In contrast, the disaggregated architecture allows for flexible resource allocation, supports heterogeneous training setups, and facilitates cross-datacenter deployment. StreamRL is designed with disaggregation from first principles and fully unlocks its potential by addressing two types of performance bottlenecks in existing disaggregated RL frameworks: pipeline bubbles, caused by stage dependencies, and skewness bubbles, resulting from long-tail output length distributions. To address pipeline bubbles, StreamRL breaks the traditional stage boundary in synchronous RL algorithms through stream generation and achieves full overlapping in asynchronous RL. To address skewness bubbles, StreamRL employs an output-length ranker model to identify long-tail samples and reduces generation time via skewness-aware dispatching and scheduling. Experiments show that StreamRL improves throughput by up to 2.66x compared to existing state-of-the-art systems, and improves cost-effectiveness by up to 1.33x in a heterogeneous, cross-datacenter setting.

--Large open-source projects receive a large number of issues (known as bugs), including software defect (i.e., bug) reports and new feature requests from their user and developer communities at a fast rate. The often limited project resources do not allow them to deal with all issues. Instead, they have to prioritize them according to the project's priorities and the issues' severities. In this paper, we propose a novel approach to automated bug prioritization based on the natural language text of the bug reports that are stored in the open bug repositories of the issue-tracking systems. We conduct topic modeling using a variant of LDA called T opicMiner-MTM and text classification with the BERT large language model to achieve a higher performance level compared to the state-of-the-art. Experimental results using an existing reference dataset containing 85,156 bug reports of the Eclipse Platform project indicate that we outperform existing approaches in terms of Accuracy, Precision, Recall, and F1-measure of the bug report priority prediction. Index T erms --automated bug prioritization, automated bug triage, mining software repositories, machine learning, natural language processing I. I NTRODUCTION Large open-source projects offer an issue-tracking system with an open bug repository, where developers and users can report the software defects they find or any new feature requests they may have. These reports are called bug reports . However, the projects' resources are limited, while processing and resolving the bug reports is typically very costly. Hence, not all bug reports in the open bug repository can be processed and handled at once.

Multimodal aspect-based sentiment classification (MASC) is an emerging task due to an increase in user-generated multimodal content on social platforms, aimed at predicting sentiment polarity toward specific aspect targets (i.e., entities or attributes explicitly mentioned in text-image pairs). Despite extensive efforts and significant achievements in existing MASC, substantial gaps remain in understanding fine-grained visual content and the cognitive rationales derived from semantic content and impressions (cognitive interpretations of emotions evoked by image content). In this study, we present Chimera: a cognitive and aesthetic sentiment causality understanding framework to derive fine-grained holistic features of aspects and infer the fundamental drivers of sentiment expression from both semantic perspectives and affective-cognitive resonance (the synergistic effect between emotional responses and cognitive interpretations). Specifically, this framework first incorporates visual patch features for patch-word alignment. Meanwhile, it extracts coarse-grained visual features (e.g., overall image representation) and fine-grained visual regions (e.g., aspect-related regions) and translates them into corresponding textual descriptions (e.g., facial, aesthetic). Finally, we leverage the sentimental causes and impressions generated by a large language model (LLM) to enhance the model's awareness of sentimental cues evoked by semantic content and affective-cognitive resonance. Experimental results on standard MASC datasets demonstrate the effectiveness of the proposed model, which also exhibits greater flexibility to MASC compared to LLMs such as GPT-4o. We have publicly released the complete implementation and dataset at https://github.com/Xillv/Chimera

Large language models (LLMs) have shown strong performance in Verilog generation from natural language description. However, ensuring the functional correctness of the generated code remains a significant challenge. This paper introduces a method that integrates verification insights from testbench into the training of Verilog generation LLMs, aligning the training with the fundamental goal of hardware design: functional correctness. The main obstacle in using LLMs for Verilog code generation is the lack of sufficient functional verification data, particularly testbenches paired with design specifications and code. To address this problem, we introduce an automatic testbench generation pipeline that decomposes the process and uses feedback from the Verilog compiler simulator (VCS) to reduce hallucination and ensure correctness. We then use the testbench to evaluate the generated codes and collect them for further training, where verification insights are introduced. Our method applies reinforcement learning (RL), specifically direct preference optimization (DPO), to align Verilog code generation with functional correctness by training preference pairs based on testbench outcomes. In evaluations on VerilogEval-Machine, VerilogEval-Human, RTLLM v1.1, RTLLM v2, and VerilogEval v2, our approach consistently outperforms state-of-the-art baselines in generating functionally correct Verilog code. We open source all training code, data, and models at https://anonymous.4open.science/r/VeriPrefer-E88B.

As large language models (LLMs) and LLM-based agents increasingly interact with humans in decision-making contexts, understanding the trust dynamics between humans and AI agents becomes a central concern. While considerable literature studies how humans trust AI agents, it is much less understood how LLM-based agents develop effective trust in humans. LLM-based agents likely rely on some sort of implicit effective trust in trust-related contexts (e.g., evaluating individual loan applications) to assist and affect decision making. Using established behavioral theories, we develop an approach that studies whether LLMs trust depends on the three major trustworthiness dimensions: competence, benevolence and integrity of the human subject. We also study how demographic variables affect effective trust. Across 43,200 simulated experiments, for five popular language models, across five different scenarios we find that LLM trust development shows an overall similarity to human trust development. We find that in most, but not all cases, LLM trust is strongly predicted by trustworthiness, and in some cases also biased by age, religion and gender, especially in financial scenarios. This is particularly true for scenarios common in the literature and for newer models. While the overall patterns align with human-like mechanisms of effective trust formation, different models exhibit variation in how they estimate trust; in some cases, trustworthiness and demographic factors are weak predictors of effective trust. These findings call for a better understanding of AI-to-human trust dynamics and monitoring of biases and trust development patterns to prevent unintended and potentially harmful outcomes in trust-sensitive applications of AI.