Current conversational AI systems often provide generic, one-size-fits-all interactions that overlook individual user characteristics and lack adaptive dialogue management. To address this gap, we introduce \textbf{HumAIne-chatbot}, an AI-driven conversational agent that personalizes responses through a novel user profiling framework. The system is pre-trained on a diverse set of GPT-generated virtual personas to establish a broad prior over user types. During live interactions, an online reinforcement learning agent refines per-user models by combining implicit signals (e.g. typing speed, sentiment, engagement duration) with explicit feedback (e.g., likes and dislikes). This profile dynamically informs the chatbot dialogue policy, enabling real-time adaptation of both content and style. To evaluate the system, we performed controlled experiments with 50 synthetic personas in multiple conversation domains. The results showed consistent improvements in user satisfaction, personalization accuracy, and task achievement when personalization features were enabled. Statistical analysis confirmed significant differences between personalized and nonpersonalized conditions, with large effect sizes across key metrics. These findings highlight the effectiveness of AI-driven user profiling and provide a strong foundation for future real-world validation.

Scientific literature question answering is a pivotal step towards new scientific discoveries. Recently, \textit{two-stage} retrieval-augmented generated large language models (RAG-LLMs) have shown impressive advancements in this domain. Such a two-stage framework, especially the second stage (reranker), is particularly essential in the scientific domain, where subtle differences in terminology may have a greatly negative impact on the final factual-oriented or knowledge-intensive answers. Despite this significant progress, the potential and limitations of these works remain unexplored. In this work, we present a Scientific Rerank-oriented RAG Benchmark (SciRerankBench), for evaluating rerankers within RAG-LLMs systems, spanning five scientific subjects. To rigorously assess the reranker performance in terms of noise resilience, relevance disambiguation, and factual consistency, we develop three types of question-context-answer (Q-C-A) pairs, i.e., Noisy Contexts (NC), Semantically Similar but Logically Irrelevant Contexts (SSLI), and Counterfactual Contexts (CC). Through systematic evaluation of 13 widely used rerankers on five families of LLMs, we provide detailed insights into their relative strengths and limitations. To the best of our knowledge, SciRerankBench is the first benchmark specifically developed to evaluate rerankers within RAG-LLMs, which provides valuable observations and guidance for their future development.

Fine-tuning massive pre-trained language models across many tasks demands adapters that are both parameter-efficient and expressive. We introduce \textbf{Kron-LoRA}, a hybrid adapter that combines Kronecker-structured factorization with low-rank LoRA compression-an integration that, to our knowledge, has not been explored in parameter-efficient fine-tuning or in matrix approximation literature. Kron-LoRA achieves up to 4$\times$ fewer parameters than standard LoRA while retaining similar expressivity. Experiments on DistilBERT, Mistral-7B, LLaMA-2-7B, and LLaMA-3-8B across eight benchmarks show that Kron-LoRA matches or exceeds LoRA baselines with modest memory savings and only a 5-8\% speed overhead. In sequential fine-tuning, it also delivers competitive cross-task transfer despite using only one-quarter of the adapter parameters. Kron-LoRA thus offers a scalable, sustainable solution for multi-task adaptation of large language models.

Recent work has shown that exoskeletons controlled through data-driven methods can dynamically adapt assistance to various tasks for healthy young adults. However, applying these methods to populations with neuromotor gait deficits, such as post-stroke hemiparesis, is challenging. This is due not only to high population heterogeneity and gait variability but also to a lack of post-stroke gait datasets to train accurate models. Despite these challenges, data-driven methods offer a promising avenue for control, potentially allowing exoskeletons to function safely and effectively in unstructured community settings. This work presents a first step towards enabling adaptive plantarflexion and dorsiflexion assistance from data-driven torque estimation during post-stroke walking. We trained a multi-task Temporal Convolutional Network (TCN) using collected data from four post-stroke participants walking on a treadmill ($R^2$ of $0.74 \pm 0.13$). The model uses data from three inertial measurement units (IMU) and was pretrained on healthy walking data from 6 participants. We implemented a wearable prototype for our ankle torque estimation approach for exoskeleton control and demonstrated the viability of real-time sensing, estimation, and actuation with one post-stroke participant.

Background and Aims: Large language models (LLMs) have shown remarkable generalization and transfer capabilities by learning from vast corpora of text and web data. Their semantic representations allow cross-task knowledge transfer and reasoning, offering promising opportunities for data-scarce and heterogeneous domains such as clinical medicine. Yet, in diagnostic tasks like sarcopenia, major challenges remain: interpretability, transparency, and deployment efficiency. Traditional machine learning (TML) models provide stable performance and feature-level attribution, ensuring traceable and auditable decision logic, but lack semantic breadth. Conversely, LLMs enable flexible inference but often function as opaque predictors. Existing integration strategies remain shallow, rarely embedding the structured reasoning of TML into LLM inference. Methods: Using sarcopenia diagnosis as a case study, SHapley Additive exPlanations (SHAP) were extracted from a baseline XGBoost model and transformed into structured, LLM-compatible representations. An actor-critic reinforcement learning (RL) strategy guided the LLM to reason over these SHAP-based inputs, producing calibrated rationales and refined decision rules. The distilled reasoning was consolidated into a structured knowledge repository and deployed via retrieval-augmented generation (RAG) for case-based inference. Results: (Omitted here.) Conclusion: By coupling SHAP-derived statistical evidence with reinforcement-trained LLM reasoning, CANDLE mitigates the interpretability-performance trade-off, enhances predictive accuracy, and preserves high decision consistency. The framework offers a scalable approach to knowledge assetization of TML models, enabling interpretable, reproducible, and clinically aligned decision support in sarcopenia and potentially broader medical domains.

Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, significantly reduce the number of trainable parameters by introducing low-rank decomposition matrices. However, existing methods perform extensive matrix multiplications in domain specialization tasks, resulting in computational inefficiency and sub-optimal fine-tuning performance. Hence, we propose LoSiA(Low-Resources Subnet Integration Adaptation), an innovative method that dynamically localizes and optimizes critical parameters during the training process. Specifically, it identifies a sub-network using gradient sparsity analysis and optimizes it as the trainable target. This design enables effective high-rank adaptation by updating only the sub-network parameters, reducing the additional matrix multiplication. We also present LoSiA-Pro, a faster implementation of LoSiA, which reduces the training latency by about $27\%$ compared to LoRA. Extensive evaluations show that our method achieves minimal performance drop compared to full fine-tuning, while requiring the least training time across domain specialization and common-sense reasoning tasks. Further analysis shows that LoSiA also reduces forgetting during continued training. The source code is available at https://github.com/KlozeWang/LoSiA.

Recent advances in deep learning have led to increasingly complex models with deeper layers and more parameters, reducing interpretability and making their decisions harder to understand. While many methods explain black-box reasoning, most lack effective interventions or only operate at sample-level without modifying the model itself. T o address this, we propose the Concept Bottleneck Model for Enhancing Human-Neural Network Mutual Understanding (CBM-HNMU). CBM-HNMU leverages the Concept Bottleneck Model (CBM) as an interpretable framework to approximate black-box reasoning and communicate conceptual understanding. Detrimental concepts are automatically identified and refined (removed/replaced) based on global gradient contributions. The modified CBM then distills corrected knowledge back into the black-box model, enhancing both interpretability and accuracy. W e evaluate CBM-HNMU on various CNN and transformer-based models across Flower-102, CIF AR-10, CIF AR-100, FGVC-Aircraft, and CUB-200, achieving a maximum accuracy improvement of 2.64% and a maximum increase in average accuracy across 1.03%.

Ensuring reliability is paramount in deep learning, particularly within the domain of medical imaging, where diagnostic decisions often hinge on model outputs. The capacity to separate out-of-distribution (OOD) samples has proven to be a valuable indicator of a model's reliability in research. In medical imaging, this is especially critical, as identifying OOD inputs can help flag potential anomalies that might otherwise go undetected. While many OOD detection methods rely on feature or logit space representations, recent works suggest these approaches may not fully capture OOD diversity. To address this, we propose a novel OOD scoring mechanism, called NERO, that leverages neuron-level relevance at the feature layer. Specifically, we cluster neuron-level relevance for each in-distribution (ID) class to form representative centroids and introduce a relevance distance metric to quantify a new sample's deviation from these centroids, enhancing OOD separability. Additionally, we refine performance by incorporating scaled relevance in the bias term and combining feature norms. Our framework also enables explainable OOD detection.

Although the power of LLM tool-use agents has ignited a flurry of recent research in this area, the curation of tool-use training data remains an open problem$-$especially for online RL training. Existing approaches to synthetic tool-use data generation tend to be non-interactive, and/or non-compositional. We introduce RandomWorld, a pipeline for the procedural generation of interactive tools and compositional tool-use data. We show that models tuned via SFT and RL on synthetic RandomWorld data improve on a range of tool-use benchmarks, and set the new SoTA for two metrics on the NESTFUL dataset. Further experiments show that downstream performance scales with the amount of RandomWorld-generated training data, opening up the possibility of further improvement through the use of entirely synthetic data.

Recent reasoning models show the ability to reflect, backtrack, and self-validate their reasoning, which is crucial in spotting mistakes and arriving at accurate solutions. A natural question that arises is how effectively models can perform such self-reevaluation. We tackle this question by investigating how well reasoning models identify and recover from four types of unhelpful thoughts: uninformative rambling thoughts, thoughts irrelevant to the question, thoughts misdirecting the question as a slightly different question, and thoughts that lead to incorrect answers. We show that models are effective at identifying most unhelpful thoughts but struggle to recover from the same thoughts when these are injected into their thinking process, causing significant performance drops. Models tend to naively continue the line of reasoning of the injected irrelevant thoughts, which showcases that their self-reevaluation abilities are far from a general "meta-cognitive" awareness. Moreover, we observe non/inverse-scaling trends, where larger models struggle more than smaller ones to recover from short irrelevant thoughts, even when instructed to reevaluate their reasoning. We demonstrate the implications of these findings with a jailbreak experiment using irrelevant thought injection, showing that the smallest models are the least distracted by harmful-response-triggering thoughts. Overall, our findings call for improvement in self-reevaluation of reasoning models to develop better reasoning and safer systems.