While many advanced LLMs are designed to handle long sequence data, we can still observe notable quality degradation even within the sequence limit. In this work, we introduce a novel approach called Scaling to Emphasize Attention for Long-context retrieval (SEAL), which enhances the retrieval performance of large language models (LLMs) over long contexts. We observe that specific attention heads are closely tied to long-context retrieval, showing positive or negative correlation with retrieval scores, and adjusting the strength of these heads boosts the quality of LLMs in long context by a large margin. Built on this insight, we propose a learning-based mechanism that leverages generated data to emphasize these heads. By applying SEAL, we achieve significant improvements in long-context retrieval performance across various tasks and models. Additionally, when combined with existing training-free context extension techniques, SEAL extends the contextual limits of LLMs while maintaining highly reliable outputs.

To mitigate the hallucination problem in large language models, DoLa exploits early exit logits from the same model as a contrastive prior. However, we found that these early exit logits tend to be flat, low in magnitude, and fail to reflect meaningful contrasts. To address this, we propose PruneCD, a novel contrastive decoding method that constructs the amateur model via layer pruning rather than early exit. This design leads to more informative and well-aligned logits, enabling more effective contrastive decoding. Through qualitative and quantitative analyses, we demonstrate that PruneCD consistently improves factuality with minimal inference overhead, offering a robust and practical approach to mitigating hallucinations in LLMs.

To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints. We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that assigns layer-wise quantization bit-widths to optimally balance model quality and memory usage. However, the combinatorial search space, with over 10¹⁰⁰ possible configurations, makes conventional black-box optimization infeasible. AMQ overcomes this challenge through four key innovations: (1) **search space pruning** using prior knowledge to exclude unpromising configurations, (2) **quantization proxy** to bypass costly format conversions during search, (3) **quality predictor** to minimize evaluation overhead, and (4) **iterative search-and-update** strategy for fast and stable convergence. By integrating these components, AMQ efficiently explores the quality–efficiency landscape, reaching the Pareto frontier and yielding LLMs that are both compact and high-performing.

Recently, leveraging reinforcement learning (RL) to fine-tune language models (LMs), known as reinforcement learning from human feedback (RLHF), has become an important research topic. However, there is still a lack of theoretical understanding of how RLHF works, the conditions under which it succeeds or fails, and whether it guarantees optimization of both likelihood 𝛽(⋅) and reward R(⋅) objectives. To address these issues, we consider RLHF as a bi-objective problem that has the nature of a Pareto optimization, present a Pareto improvement condition that is necessary to obtain Pareto-efficient policies, and propose a simple yet powerful method named reward dropout that guarantees a Pareto improvement. To demonstrate the performance of reward dropout, two benchmark datasets commonly used in text style transfer tasks were utilized in our study: sentiment and topic datasets sourced from Yelp and AG_News, respectively. Our experiments highlight that paying attention to a few samples with higher rewards leads to greater Pareto improvements regardless of model size. We also demonstrate that the effect of reward dropout is generalizable and most effective with non-pretrained target models, saving the effort of pretraining.

With the rapid growth in the use of fine-tuning for large language models (LLMs), optimizing fine-tuning while keeping inference efficient has become highly important. However, this is a challenging task as it requires improvements in all aspects, including inference speed, fine-tuning speed, memory consumption, and, most importantly, model quality. Previous studies have attempted to achieve this by combining quantization with fine-tuning, but they have failed to enhance all four aspects simultaneously. In this study, we propose a new lightweight technique called Quantization for Efficient Fine-Tuning (QEFT). QEFT accelerates both inference and fine-tuning, is supported by robust theoretical foundations, offers high flexibility, and maintains good hardware compatibility. Our extensive experiments demonstrate that QEFT matches the quality and versatility of full-precision parameter-efficient fine-tuning, while using fewer resources. Our code is available at https://github.com/xvyaward/qeft.

Generating as diverse molecules as possible with desired properties is crucial for drug discovery research, which invokes many approaches based on deep generative models today. Despite recent advancements in these models, particularly in variational autoencoders (VAEs), generative adversarial networks (GANs), Transformers, and diffusion models, a significant challenge known as the sample bias problem remains. This problem occurs when generated molecules targeting the same protein tend to be structurally similar, reducing the diversity of generation. To address this, we propose leveraging multi-hop relationships among proteins and compounds. Our model, Repurformer, integrates bi-directional pretraining with Fast Fourier Transform (FFT) and low-pass filtering (LPF) to capture complex interactions and generate diverse molecules. A series of experiments on BindingDB dataset confirm that Repurformer successfully creates substitutes for anchor compounds that resemble positive compounds, increasing diversity between the anchor and generated compounds.