Related papers: Flexible Realignment of Language Models

Low-rank adaptation (LoRA) has been developed as an efficient approach for adapting large language models (LLMs) by fine-tuning two low-rank matrices, thereby reducing the number of trainable parameters. However, prior research indicates…

The high inference demands of transformer-based Large Language Models (LLMs) pose substantial challenges in their deployment. To this end, we introduce Neural Block Linearization (NBL), a novel framework for accelerating transformer model…

Using back-propagation and its variants to train deep networks is often problematic for new users. Issues such as exploding gradients, vanishing gradients, and high sensitivity to weight initialization strategies often make networks…

We propose Quantum-informed Tensor Adaptation (QuanTA), a novel, easy-to-implement, fine-tuning method with no inference overhead for large-scale pre-trained language models. By leveraging quantum-inspired methods derived from quantum…

Large language model (LLM) reinforcement learning (RL) pipelines are often bottlenecked by rollout generation, making end-to-end training slow. Recent work mitigates this by running rollouts with quantization to accelerate decoding, which…

Low-Rank Adaptation (LoRA) has significantly advanced parameter-efficient fine-tuning of large pretrained models. LoRA augments the pre-trained weights of a model by adding the product of two smaller matrices that together form a low-rank…

Although Large Language Models (LLMs) have shown promise for human-like conversations, they are primarily pre-trained on text data. Incorporating audio or video improves performance, but collecting large-scale multimodal data and…

Large Language Models have shown remarkable capabilities in the NLP domain. Their effectiveness can mainly be attributed to their ability to adapt to an array of downstream tasks. However, generally, full fine-tuning is a computationally…

Pre-trained Large Language Models (LLMs) have revolutionized text processing, yet adapting Transformer-based neural networks to non-textual scientific modalities typically requires specialized architectures and extensive computational…

Diffusion language models (DLMs) have recently demonstrated capabilities that complement standard autoregressive (AR) models, particularly in non-sequential generation and bidirectional editing. Although recent work has shown that…

To mitigate reward hacking from response verbosity, modern preference optimization methods are increasingly adopting length normalization (e.g., SimPO, ORPO, LN-DPO). While effective against this bias, we demonstrate that length…

The widespread utilization of language models in modern applications is inconceivable without Parameter-Efficient Fine-Tuning techniques, such as low-rank adaptation ($\texttt{LoRA}$), which adds trainable adapters to selected layers.…

Large language models (LLMs) exhibit cultural bias from overrepresented viewpoints in training data, yet cultural alignment remains a challenge due to limited cultural knowledge and a lack of exploration into effective learning approaches.…

The advent of large language models (LLMs) has revolutionized natural language processing, enabling unprecedented capabilities in understanding and generating human-like text. However, the computational cost and convergence times associated…

Adapter-based Federated Large Language Models (FedLLMs) are widely adopted to reduce the computational, storage, and communication overhead of full-parameter fine-tuning for web-scale applications while preserving user privacy. By freezing…

Large Language Model (LLM) unlearning aims to remove targeted knowledge from a trained model, but practical deployments often require post-training quantization (PTQ) for efficient inference. However, aggressive low-bit PTQ can mask…

Transformer-based large language models (LLMs) have achieved remarkable success across various tasks. Yet, fine-tuning such massive models in federated learning (FL) settings poses significant challenges due to resource constraints and…

Large language models (LLMs) commonly employ autoregressive generation during inference, leading to high memory bandwidth demand and consequently extended latency. To mitigate this inefficiency, we present Bi-directional Tuning for lossless…

Low Rank Adaptation (LoRA) has gained massive attention in the recent generative AI research. One of the main advantages of LoRA is its ability to be fused with pretrained models, adding no overhead during inference. However, from a mobile…

Large language models (LLMs) exhibit persistent miscalibration, especially after instruction tuning and preference alignment. Modified training objectives can improve calibration, but retraining is expensive. Inference-time steering offers…