相关论文: Steering Information Utility in Key-Value Memory f…

The mechanisms by which reasoning training reshapes LLMs' internal computations remain unclear. We study lightweight steering vectors inserted into the base model's residual stream and trained with a reinforcement-learning objective. These…

Large language models (LLMs) have recently emerged as powerful tools for tackling many language-processing tasks. Despite their success, training and fine-tuning these models is still far too computationally and memory intensive. In this…

Large language models (LLMs) achieve remarkable advancements by leveraging tools to interact with environments, a critical step toward generalized AI. However, the standard supervised fine-tuning (SFT) approach, which relies on large-scale…

AfterLearnER (After Learning Evolutionary Retrofitting) consists in applying evolutionary optimization to refine fully trained machine learning models by optimizing a set of carefully chosen parameters or hyperparameters of the model, with…

Recent progress in Multimodal Large Language Models (MLLMs) has unlocked powerful cross-modal reasoning abilities, but also raised new safety concerns, particularly when faced with adversarial multimodal inputs. To improve the safety of…

Large language models (LLMs) have revolutionized natural language processing (NLP) by excelling at understanding and generating human-like text. However, their widespread deployment can be prohibitively expensive. SortedNet is a recent…

Language model post-training has enhanced instruction-following and performance on many downstream tasks, but also comes with an often-overlooked cost on tasks with many possible valid answers. On many tasks such as creative writing,…

Existing methods fail to effectively steer Large Language Models (LLMs) between textual reasoning and code generation, leaving symbolic computing capabilities underutilized. We introduce CodeSteer, an effective method for guiding LLM…

Controlling multiple behavioral attributes in large language models (LLMs) at inference time is a challenging problem due to interference between attributes and the limitations of linear steering methods, which assume additive behavior in…

Steering vectors have emerged as a lightweight and effective approach for aligning large language models (LLMs) at inference time, enabling modulation over model behaviors by shifting LLM representations towards a target behavior. However,…

Large Language Models (LLMs) encode vast amounts of parametric knowledge during pre-training. As world knowledge evolves, effective deployment increasingly depends on their ability to faithfully follow externally retrieved context. When…

Large Language Models (LLMs), despite advances in instruction tuning, often fail to follow complex user instructions. Activation steering techniques aim to mitigate this by manipulating model internals, but have a potential risk of…

Large language models are classically trained in stages: pretraining on raw text followed by post-training for instruction following and reasoning. However, this separation creates a fundamental limitation: many desirable behaviors such as…

Discrete Diffusion Language Models (DLMs) offer a promising non-autoregressive alternative for text generation, yet effective mechanisms for inference-time control remain relatively underexplored. Existing approaches include sampling-level…

Supervised fine-tuning (SFT) provides the standard approach for teaching LLMs new behaviors from offline expert demonstrations. However, standard SFT uniformly fits all samples -- including those with low likelihood under the base model --…

The deployment of large language models (LLMs) in real-world applications is increasingly limited by their high inference cost. While recent advances in dynamic token-level computation allocation attempt to improve efficiency by selectively…

Pre-trained vision-language models provide a robust foundation for efficient transfer learning across various downstream tasks. In the field of video action recognition, mainstream approaches often introduce additional modules to capture…

Large language models (LLMs) excel at diverse tasks, but their deployment on resource-constrained devices remains challenging. Existing methods like quantization, pruning, and distillation can reduce memory footprint but often demand…

The memorization of training data by Large Language Models (LLMs) poses significant risks, including privacy leaks and the regurgitation of copyrighted content. Activation steering, a technique that directly intervenes in model activations,…

Steering methods influence Large Language Model behavior by identifying semantic directions in hidden representations, but are typically realized through inference-time activation interventions that apply a fixed, global modification to the…