Related papers: Steering Llama 2 via Contrastive Activation Additi…

Audio diffusion models can synthesize high-fidelity music from text, yet achieving fine-grained control over specific musical attributes remains challenging, as their internal mechanisms for representing high-level concepts are poorly…

We present CFAAR, a program repair assistance technique that operates by selectively altering the outcome of suspicious predicates in order to yield expected behavior. CFAAR is applicable to defects that are repairable by negating…

Providing high-quality feedback to Large Language Models (LLMs) on a diverse training distribution can be difficult and expensive, and providing feedback only on a narrow distribution can result in unintended generalizations. To better…

Given the prompt "Rome is in", can we steer a language model to flip its prediction of an incorrect token "France" to a correct token "Italy" by only multiplying a few relevant activation vectors with scalars? We argue that successfully…

Large language models (LLMs) enable conversational agents (CAs) to express distinctive personalities, raising new questions about how such designs shape user perceptions. This study investigates how personality expression levels and…

Steering has emerged as a practical approach to enable post-hoc guidance of LLMs towards enforcing a specific behavior. However, it remains largely underexplored for multimodal LLMs (MLLMs); furthermore, existing steering techniques, such…

Pre-trained Language Models are widely used in many important real-world applications. However, recent studies show that these models can encode social biases from large pre-training corpora and even amplify biases in downstream…

Vision-Language-Action systems follow instructions to execute multi-step tasks in multimodal environments. Recent VLA approaches typically rely on post-hoc correction mechanisms or operate under fixed task decompositions and alignment…

Large Audio-Language Models and Multi-Modal Large Language Models have demonstrated strong capabilities in tasks such as Audio Question Answering (AQA), Audio Captioning, and Automatic Speech Recognition (ASR). However, there is growing…

When language model agents tackle complex software engineering tasks, they often degrade over long trajectories, which we define as *agent drift*. We focus on two recurring failure modes *overthinking* and *overacting*, i.e., where the…

Activation steering methods control large language model (LLM) behavior by modifying internal activations at inference time. However, most existing activation steering methods rely on a fixed steering strength, leading to either…

Despite extensive efforts in safety alignment, large language models (LLMs) remain vulnerable to jailbreak attacks. Activation steering offers a training-free defense method but relies on fixed steering coefficients, resulting in suboptimal…

Creating personalized and adaptable conversational AI remains a key challenge. This paper introduces a Continuous Learning Conversational AI (CLCA) approach, implemented using A2C reinforcement learning, to move beyond static Large Language…

Recent reasoning-augmented Vision-Language-Action (VLA) models have improved the interpretability of end-to-end autonomous driving by generating intermediate reasoning traces. Yet these models primarily describe what they perceive and…

Conversational agents struggle to handle long conversations due to context window limitations. Therefore, memory systems are developed to leverage essential historical information. Existing memory systems typically follow a pipeline of…

As large language models (LLMs) move into persistent, user-facing roles, their behavior must be understood not as isolated responses but as a trajectory unfolding over sustained interaction. We introduce the concept of the chain-of-affect…

We propose Instruct2Attack (I2A), a language-guided semantic attack that generates semantically meaningful perturbations according to free-form language instructions. We make use of state-of-the-art latent diffusion models, where we…

Latent Action Models (LAMs) have rapidly gained traction as an important component in the pre-training pipelines of leading Vision-Language-Action models. However, they fail when observations contain action-correlated distractors, often…

Responsible deployment of language models requires mechanisms for refusing unsafe prompts while preserving model performance. While most approaches modify model weights through additional training, we explore an alternative: steering model…

LLMs' remarkable ability to tackle a wide range of language tasks opened new opportunities for collaborative human-AI problem solving. LLMs can amplify human capabilities by applying their intuitions and reasoning strategies at scale. We…