Large Language Models require substantial computational resources for inference, posing deployment challenges. While dynamic pruning offers superior efficiency over static methods through adaptive circuit selection, it exacerbates alignment degradation by retaining only input-dependent safety-critical circuit preservation across diverse inputs. As a result, addressing these heightened alignment vulnerabilities remains critical. We introduce Alignment-Aware Probe Pruning (AAPP), a dynamic structured pruning method that adaptively preserves alignment-relevant circuits during inference, building upon Probe Pruning. Experiments on LLaMA 2-7B, Qwen2.5-14B-Instruct, and Gemma-3-12B-IT show AAPP improves refusal rates by 50\% at matched compute, enabling efficient yet safety-preserving LLM deployment.

This study investigates embedding reconstruction attacks in large language models (LLMs) applied to genomic sequences, with a specific focus on how fine-tuning affects vulnerability to these attacks. Building upon Pan et al.'s seminal work demonstrating that embeddings from pretrained language models can leak sensitive information, we conduct a comprehensive analysis using the HS3D genomic dataset to determine whether task-specific optimization strengthens or weakens privacy protections. Our research extends Pan et al.'s work in three significant dimensions. First, we apply their reconstruction attack pipeline to pretrained and fine-tuned model embeddings, addressing a critical gap in their methodology that did not specify embedding types. Second, we implement specialized tokenization mechanisms tailored specifically for DNA sequences, enhancing the model's ability to process genomic data, as these models are pretrained on natural language and not DNA. Third, we perform a detailed comparative analysis examining position-specific, nucleotide-type, and privacy changes between pretrained and fine-tuned embeddings. We assess embeddings vulnerabilities across different types and dimensions, providing deeper insights into how task adaptation shifts privacy risks throughout genomic sequences. Our findings show a clear distinction in reconstruction vulnerability between pretrained and fine-tuned embeddings. Notably, fine-tuning strengthens resistance to reconstruction attacks in multiple architectures -- XLNet (+19.8\%), GPT-2 (+9.8\%), and BERT (+7.8\%) -- pointing to task-specific optimization as a potential privacy enhancement mechanism. These results highlight the need for advanced protective mechanisms for language models processing sensitive genomic data, while highlighting fine-tuning as a potential privacy-enhancing technique worth further exploration.

With the widespread application of large language models (LLMs) in various fields, the security challenges they face have become increasingly prominent, especially the issue of jailbreak. These attacks induce the model to generate erroneous or uncontrolled outputs through crafted inputs, threatening the generality and security of the model. Although existing defense methods have shown some effectiveness, they often struggle to strike a balance between model generality and security. Excessive defense may limit the normal use of the model, while insufficient defense may lead to security vulnerabilities. In response to this problem, we propose a Knowledge Graph Defense Framework (KG-DF). Specifically, because of its structured knowledge representation and semantic association capabilities, Knowledge Graph(KG) can be searched by associating input content with safe knowledge in the knowledge base, thus identifying potentially harmful intentions and providing safe reasoning paths. However, traditional KG methods encounter significant challenges in keyword extraction, particularly when confronted with diverse and evolving attack strategies. To address this issue, we introduce an extensible semantic parsing module, whose core task is to transform the input query into a set of structured and secure concept representations, thereby enhancing the relevance of the matching process. Experimental results show that our framework enhances defense performance against various jailbreak attack methods, while also improving the response quality of the LLM in general QA scenarios by incorporating domain-general knowledge.

Large language models exhibit a peculiar epistemic pathology: they speak as if they know, even when they do not. This paper argues that such confident fabrication, what I call the polite liar, is a structural consequence of reinforcement learning from human feedback (RLHF). Building on Frankfurt's analysis of bullshit as communicative indifference to truth, I show that this pathology is not deception but structural indifference: a reward architecture that optimizes for perceived sincerity over evidential accuracy. Current alignment methods reward models for being helpful, harmless, and polite, but not for being epistemically grounded. As a result, systems learn to maximize user satisfaction rather than truth, performing conversational fluency as a virtue. I analyze this behavior through the lenses of epistemic virtue theory, speech-act philosophy, and cognitive alignment, showing that RLHF produces agents trained to mimic epistemic confidence without access to epistemic justification. The polite liar thus reveals a deeper alignment tension between linguistic cooperation and epistemic integrity. The paper concludes with an "epistemic alignment" principle: reward justified confidence over perceived fluency.

We introduce Motif-2-12.7B, a new open-weight foundation model that pushes the efficiency frontier of large language models by combining architectural innovation with system-level optimization. Designed for scalable language understanding and robust instruction generalization under constrained compute budgets, Motif-2-12.7B builds upon Motif-2.6B with the integration of Grouped Differential Attention (GDA), which improves representational efficiency by disentangling signal and noise-control attention pathways. The model is pre-trained on 5.5 trillion tokens spanning diverse linguistic, mathematical, scientific, and programming domains using a curriculum-driven data scheduler that gradually changes the data composition ratio. The training system leverages the MuonClip optimizer alongside custom high-performance kernels, including fused PolyNorm activations and the Parallel Muon algorithm, yielding significant throughput and memory efficiency gains in large-scale distributed environments. Post-training employs a three-stage supervised fine-tuning pipeline that successively enhances general instruction adherence, compositional understanding, and linguistic precision. Motif-2-12.7B demonstrates competitive performance across diverse benchmarks, showing that thoughtful architectural scaling and optimized training design can rival the capabilities of much larger models.

Current evaluations of LLMs for code generation emphasize functional correctness, overlooking the fact that functionally correct solutions can differ significantly in algorithmic complexity. For instance, an $(O(n^2))$ versus $(O(n \log n))$ sorting algorithm may yield similar output but incur vastly different performance costs in production. This discrepancy reveals a critical limitation in current evaluation methods: they fail to capture the behavioral and performance diversity among correct solutions. To address this, we introduce a principled framework for evaluating the dynamic stability of generated code. We propose two metrics derived from opcode distributions: Static Canonical Trace Divergence (SCTD), which captures algorithmic structure diversity across generated solutions, and Dynamic Canonical Trace Divergence (DCTD), which quantifies runtime behavioral variance. Their ratio, the Behavioral Expression Factor (BEF), serves as a diagnostic signal: it indicates critical runtime instability when BEF $\ll$ 1 and functional redundancy when BEF $\gg$ 1. Empirical results on BigOBench and CodeContests show that state-of-the-art LLMs exhibit significant algorithmic variance even among functionally correct outputs. Notably, increasing sampling temperature improves pass@1 rates but degrades stability, revealing an unrecognized trade-off: searching for correct solutions in diverse output spaces introduces a "penalty of instability" between correctness and behavioral consistency. Our findings call for stability-aware objectives in code generation and new benchmarks with asymptotic test cases for robust, real-world LLM evaluation.

This paper introduces DuTerm, a novel two-stage architecture for terminology-constrained machine translation. Our system combines a terminology-aware NMT model, adapted via fine-tuning on large-scale synthetic data, with a prompt-based LLM for post-editing. The LLM stage refines NMT output and enforces terminology adherence. We evaluate DuTerm on English-to German, English-to-Spanish, and English-to-Russian with the WMT 2025 Terminology Shared Task corpus. We demonstrate that flexible, context-driven terminology handling by the LLM consistently yields higher quality translations than strict constraint enforcement. Our results highlight a critical trade-off, revealing that an LLM's work best for high-quality translation as context-driven mutators rather than generators.

Evaluating log summarization systems is challenging due to the lack of high-quality reference summaries and the limitations of existing metrics like ROUGE and BLEU, which depend on surface-level lexical overlap. We introduce REFLEX, a reference-free evaluation metric for log summarization based on large language model (LLM) judgment. REFLEX uses LLMs as zero-shot evaluators to assess summary quality along dimensions such as relevance, informativeness, and coherence, without requiring gold-standard references or human annotations. We show that REFLEX produces stable, interpretable, and fine-grained evaluations across multiple log summarization dataset, and more effectively distinguishes model outputs than traditional metrics. REFLEX provides a scalable alternative for evaluating log summaries in real-world settings where reference data is scarce or unavailable.

Large Language Models (LLMs) have demonstrated remarkable capabilities in modeling sequential textual data and generalizing across diverse tasks. However, adapting LLMs to effectively handle structural data, such as knowledge graphs or web data, remains a challenging problem. Some approaches adopt complex strategies to convert graphs into text sequences, resulting in significant token overhead and rendering them impractical for large-scale graphs. Others introduce additional modules to encode graphs into fixed-size token representations for LLMs. However, these methods typically require large-scale post-training on graph-text corpus and complex alignment procedures, yet often yield sub-optimal results due to poor modality alignment. Inspired by in-parameter knowledge injection for test-time adaptation of LLMs, we propose GRIP, a novel framework that equips LLMs with the ability to internalize complex relational information from graphs through carefully designed fine-tuning tasks. This knowledge is efficiently stored within lightweight LoRA parameters, enabling the fine-tuned LLM to perform a wide range of graph-related tasks without requiring access to the original graph at inference time. Extensive experiments across multiple benchmarks validate the effectiveness and efficiency of our approach.

This study explores whether large language models (LLMs) can simulate valid student responses for educational measurement. Using GPT - 4o, 2000 virtual student personas were generated. Each persona completed the Academic Motivation Scale (AMS). Factor analys es(EFA and CFA) and clustering showed GPT - 4o reproduced the AMS structure and distinct motivational subgroups.