Built on Llama 3.1-8B and scaled to 10B parameters, Llama-GENBA-10B is continuously pretrained on 164B tokens (82B English, 82B German, and 80M Bavarian), balancing resources while preventing English dominance. Targeted at the German NLP community, the model also promotes Bavarian as a low-resource language. Development tackled four challenges: (1) curating a multilingual corpus despite Bavarian scarcity, (2) creating a unified tokenizer for English, German, and Bavarian, (3) optimizing architecture and language-ratio hyperparame-ters for cross-lingual transfer, and (4) establishing the first standardized trilingual evaluation suite by translating German benchmarks into Bavarian. Evaluations show that Llama-GENBA-10B achieves strong cross-lingual performance, with the fine-tuned variant surpassing Apertus-8B-2509 and gemma-2-9b in Bavarian and establishing itself as the best model in its class for this language, while also outperforming EuroLLM in English and matching its results in German. Training on the Cerebras CS-2 demonstrated efficient large-scale multilingual pretraining with documented energy use, offering a blueprint for inclusive foundation models that integrate low-resource languages.

Abstract--Intent detection is a crucial component of modern conversational systems, since accurately identifying user intent at the beginning of a conversation is essential for generating effective responses. Currently, most of the intent detectors can only work effectively with the assumption that high-quality labeled data is available (i.e., the collected data is labeled by domain experts). T o ease this process, recent efforts have focused on studying this problem under a more challenging few-shot scenario. These approaches primarily leverage large-scale unlabeled dialogue text corpora to pretrain language models through various pretext tasks, followed by fine-tuning for intent detection with very limited annotations. Despite the improvements achieved, existing methods have predominantly focused on textual data, neglecting to effectively capture the crucial structural information inherent in conversational systems, such as the query-query relation and query-answer relation. Specifically, the query-query relation captures the semantic relevance between two queries within the same session, reflecting the user's refinement of her request, while the query-answer relation represents the conversational agent's clarification and response to a user query . T o address this gap, we propose SAID, a novel framework that integrates both textual and relational structure information in a unified manner for model pretraining for the first time. Firstly, we introduce a relation-aware prompt module, which employs learnable relation tokens as soft prompts, enabling the model to learn shared knowledge across multiple relations and become explicitly aware of how to interpret query text within the context of these relations. Secondly, we reformulate the few-shot intent detection problem using prompt learning by creating a new intent-specific relation-aware prompt, which incorporates intent-specific relation tokens alongside the semantic information embedded in intent names, helping the pretrained model effectively transfer the pretrained knowledge acquired from related relational perspectives. Building on this framework, we further propose a novel mechanism, the query-adaptive attention network (QueryAdapt), which operates at the relation token level by generating intent-specific relation tokens from well-learned query-query and query-answer relations explicitly, enabling more fine-grained knowledge transfer. Extensive experimental results on two real-world datasets demonstrate that SAID significantly outperforms state-of-the-art methods, achieving improvements of up to 27% in the 3-shot setting. When equipped with the relation token-level QueryAdapt module, it yields additional performance gains of up to 21% in the same setting.

The emotional content of song lyrics plays a pivotal role in shaping listener experiences and influencing musical preferences. This paper investigates the task of multi-label emotional attribution of song lyrics by predicting six emotional intensity scores corresponding to six fundamental emotions. A manually labeled dataset is constructed using a mean opinion score (MOS) approach, which aggregates annotations from multiple human raters to ensure reliable ground-truth labels. Leveraging this dataset, we conduct a comprehensive evaluation of several publicly available large language models (LLMs) under zero-shot scenarios. Additionally, we fine-tune a BERT-based model specifically for predicting multi-label emotion scores. Experimental results reveal the relative strengths and limitations of zero-shot and fine-tuned models in capturing the nuanced emotional content of lyrics. Our findings highlight the potential of LLMs for emotion recognition in creative texts, providing insights into model selection strategies for emotion-based music information retrieval applications. The labeled dataset is available at https://github.com/LLM-HITCS25S/LyricsEmotionAttribution.

The widespread deployment of LLMs across enterprise services has created a critical security blind spot. Organizations operate multiple LLM services handling billions of queries daily, yet regulatory compliance boundaries prevent these services from sharing threat intelligence about prompt injection attacks, the top security risk for LLMs. When an attack is detected in one service, the same threat may persist undetected in others for months, as privacy regulations prohibit sharing user prompts across compliance boundaries. We present BinaryShield, the first privacy-preserving threat intelligence system that enables secure sharing of attack fingerprints across compliance boundaries. BinaryShield transforms suspicious prompts through a unique pipeline combining PII redaction, semantic embedding, binary quantization, and randomized response mechanism to potentially generate non-invertible fingerprints that preserve attack patterns while providing privacy. Our evaluations demonstrate that BinaryShield achieves an F1-score of 0.94, significantly outperforming SimHash (0.77), the privacy-preserving baseline, while achieving 64x storage reduction and 38x faster similarity search compared to dense embeddings.

Large Language Models (LLMs) require high quality preference datasets to align with human preferences. However, conventional methods for constructing such datasets face significant challenges: reliance on pre-collected instructions often leads to distribution mismatches with target models, while the need for sampling multiple stochastic responses introduces substantial computational overhead. In this work, we explore a paradigm shift by leveraging inherent regulation of LLMs' representation space for efficient and tailored preference dataset construction, named Icon$^{2}$. Specifically, it first extracts layer-wise direction vectors to encode sophisticated human preferences and then uses these vectors to filter self-synthesized instructions based on their inherent consistency. During decoding, bidirectional inherent control is applied to steer token representations, enabling the precise generation of response pairs with clear alignment distinctions. Experimental results demonstrate significant improvements in both alignment and efficiency. Llama3-8B and Qwen2-7B achieve an average win rate improvement of 13.89% on AlpacaEval 2.0 and 13.45% on Arena-Hard, while reducing computational costs by up to 48.1%.

Foundation models face growing compute and memory bottlenecks, hindering deployment on resource-limited platforms. While compression techniques such as pruning and quantization are widely used, most rely on uniform heuristics that ignore architectural and runtime heterogeneity. Profiling tools expose per-layer latency, memory, and compute cost, yet are rarely integrated into automated pipelines. We propose ProfilingAgent, a profiling-guided, agentic approach that uses large language models (LLMs) to automate compression via structured pruning and post-training dynamic quantization. Our modular multi-agent system reasons over static metrics (MACs, parameter counts) and dynamic signals (latency, memory) to design architecture-specific strategies. Unlike heuristic baselines, ProfilingAgent tailors layer-wise decisions to bottlenecks. Experiments on ImageNet-1K, CIFAR-10, and CIFAR-100 with ResNet-101, ViT-B/16, Swin-B, and DeiT-B/16 show pruning maintains competitive or improved accuracy (about 1% drop on ImageNet-1K, +2% gains for ViT-B/16 on smaller datasets), while quantization achieves up to 74% memory savings with <0.5% accuracy loss. Our quantization also yields consistent inference speedups of up to 1.74 times faster. Comparative studies with GPT-4o and GPT-4-Turbo highlight the importance of LLM reasoning quality for iterative pruning. These results establish agentic systems as scalable solutions for profiling-guided model optimization.

Multimodal large language models (MLLMs) have shown strong vision-language reasoning abilities but still lack robust 3D spatial understanding, which is critical for autonomous driving. This limitation stems from two key challenges: (1) the difficulty of constructing accessible yet effective 3D representations without expensive manual annotations, and (2) the loss of fine-grained spatial details in VLMs due to the absence of large-scale 3D vision-language pretraining. To address these challenges, we propose OccVLA, a novel framework that integrates 3D occupancy representations into a unified multimodal reasoning process. Unlike prior approaches that rely on explicit 3D inputs, OccVLA treats dense 3D occupancy as both a predictive output and a supervisory signal, enabling the model to learn fine-grained spatial structures directly from 2D visual inputs. The occupancy prediction are regarded as implicit reasoning processes and can be skipped during inference without performance degradation, thereby adding no extra computational overhead. OccVLA achieves state-of-the-art results on the nuScenes benchmark for trajectory planning and demonstrates superior performance on 3D visual question-answering tasks, offering a scalable, interpretable, and fully vision-based solution for autonomous driving. Recently, end-to-end autonomous driving (Hu et al., 2022; Jiang et al., 2023; contributors, 2023; Hu et al., 2023) has witnessed remarkable advances, driven by increasing demands for real-world deployments.

The effective execution of tests for REST APIs remains a considerable challenge for development teams, driven by the inherent complexity of distributed systems, the multitude of possible scenarios, and the limited time available for test design. Exhaustive testing of all input combinations is impractical, often resulting in undetected failures, high manual effort, and limited test coverage. To address these issues, we introduce RestTSLLM, an approach that uses Test Specification Language (TSL) in conjunction with Large Language Models (LLMs) to automate the generation of test cases for REST APIs. The approach targets two core challenges: the creation of test scenarios and the definition of appropriate input data. The proposed solution integrates prompt engineering techniques with an automated pipeline to evaluate various LLMs on their ability to generate tests from OpenAPI specifications. The evaluation focused on metrics such as success rate, test coverage, and mutation score, enabling a systematic comparison of model performance. The results indicate that the best-performing LLMs - Claude 3.5 Sonnet (Anthropic), Deepseek R1 (Deepseek), Qwen 2.5 32b (Alibaba), and Sabia 3 (Maritaca) - consistently produced robust and contextually coherent REST API tests. Among them, Claude 3.5 Sonnet outperformed all other models across every metric, emerging in this study as the most suitable model for this task. These findings highlight the potential of LLMs to automate the generation of tests based on API specifications.

This work presents a Biomedical Literature Question Answering (Q&A) system based on a Retrieval-Augmented Generation (RAG) architecture, designed to improve access to accurate, evidence-based medical information. Addressing the shortcomings of conventional health search engines and the lag in public access to biomedical research, the system integrates diverse sources, including PubMed articles, curated Q&A datasets, and medical encyclopedias ,to retrieve relevant information and generate concise, context-aware responses. The retrieval pipeline uses MiniLM-based semantic embeddings and FAISS vector search, while answer generation is performed by a fine-tuned Mistral-7B-v0.3 language model optimized using QLoRA for efficient, low-resource training. The system supports both general medical queries and domain-specific tasks, with a focused evaluation on breast cancer literature demonstrating the value of domain-aligned retrieval. Empirical results, measured using BERTScore (F1), show substantial improvements in factual consistency and semantic relevance compared to baseline models. The findings underscore the potential of RAG-enhanced language models to bridge the gap between complex biomedical literature and accessible public health knowledge, paving the way for future work on multilingual adaptation, privacy-preserving inference, and personalized medical AI systems.

Reinforcement learning with verifiable rewards has significantly advanced reasoning in large language models (LLMs), but such signals remain coarse, offering only binary correctness feedback. This limitation often results in inefficiencies, including overly verbose reasoning and high computational cost, while existing solutions often compromise accuracy. To address this, we introduce self-aligned reward (SAR), a self-guided signal that complements verifiable rewards to encourage both reasoning accuracy and efficiency. SAR is defined as the relative perplexity difference between an answer conditioned on the query and the standalone answer, thereby favoring responses that are concise and query-specific. Quantitative analysis reveals that SAR reliably distinguishes answer quality: concise, correct answers score higher than redundant ones, and partially correct answers score higher than entirely incorrect ones. Evaluation on 4 models across 7 benchmarks shows that integrating SAR with prevalent RL algorithms like PPO and GRPO improves accuracy by 4%, while reducing inference cost by 30%. Further analysis demonstrates that SAR achieves a Pareto-optimal trade-off between correctness and efficiency compared to reward signals based on length or self-confidence. We also show that SAR shortens responses while preserving advanced reasoning behaviors, demonstrating its ability to suppress unnecessary elaboration without losing critical reasoning. These results highlight the promise of self-aligned reward as a fine-grained complement to verifiable rewards, paving the way for more efficient and effective LLM training.Figure 1: Training with self-aligned reward enhances both efficiency and accuracy. We present the relative gains in efficiency and accuracy compared to the respective base model in math reasoning benchmarks. Efficiency gain is measured as the drop in average response length. Recently, reinforcement learning (RL) with verifiable rewards has attracted broad attention in LLM training, demonstrating remarkable improvements in reasoning skills (Guo et al., 2025; Jaech et al., 2024). Work done during an internship at Amazon Web Services.