相关论文: Adaptive Multi-Subspace Representation Steering fo…

Large Multi-Modal Models (LMMs) have demonstrated impressive capabilities as general-purpose chatbots able to engage in conversations about visual inputs. However, their responses are influenced by societal biases present in their training…

Effective contact-rich manipulation requires robots to synergistically leverage vision, force, and proprioception. However, Reinforcement Learning agents struggle to learn in such multisensory settings, especially amidst sensory noise and…

Activation-based control steers large language models (LLMs) by intervening on their internal representations during inference, and has emerged as an effective paradigm for controlling behaviors such as persona and style. However, existing…

Language models (LMs) are increasingly used in high-stakes, multi-agent settings, where following instructions and maintaining value alignment are critical. Most alignment research focuses on interactions between a single LM and a single…

Autonomous Driving Systems (ADSs) are revolutionizing transportation by reducing human intervention, improving operational efficiency, and enhancing safety. Large Language Models (LLMs) have been integrated into ADSs to support high-level…

Streaming automatic speech recognition (ASR) aims to emit each hypothesized word as quickly and accurately as possible, while full-context ASR waits for the completion of a full speech utterance before emitting completed hypotheses. In this…

Pre-trained models have been a foundational approach in speech recognition, albeit with associated additional costs. In this study, we propose a regularization technique that facilitates the training of visual and audio-visual speech…

Modern software systems often have to cope with uncertain operation conditions, such as changing workloads or fluctuating interference in a wireless network. To ensure that these systems meet their goals these uncertainties have to be…

Adapting large language models to multiple tasks can cause cross-skill interference, where improvements for one skill degrade another. While methods such as LoRA impose orthogonality constraints at the weight level, they do not fully…

Large language models (LLMs) generate high-dimensional embeddings that capture rich semantic and syntactic information. However, high-dimensional embeddings exacerbate computational complexity and storage requirements, thereby hindering…

Large language models (LLMs) have achieved remarkable performance across many tasks, yet aligning them with desired behaviors remains challenging. Activation intervention has emerged as an effective and economical method to modify the…

Aligning pretrained audio encoders and Large Language Models (LLMs) offers a promising, parameter-efficient path to building powerful multimodal agents. However, existing methods often require costly full-model finetuning or rely on static…

In real-world applications, automatic speech recognition (ASR) systems must handle overlapping speech from multiple speakers and recognize rare words like technical terms. Traditional methods address multi-talker ASR and contextual biasing…

Self-supervised learning (SSL) methods which learn representations of data without explicit supervision have gained popularity in speech-processing tasks, particularly for single-talker applications. However, these models often have…

If large language models operate in a universal semantic space, then switching between languages should require only a simple activation offset. To test this, we take multilingual in-context learning as a case study, where few-shot…

The Active Subspace (AS) method is a widely used technique for identifying the most influential directions in high-dimensional input spaces that affect the output of a computational model. The standard AS algorithm requires a sufficient…

Automatic speech recognition (ASR) has benefited from advances in pretrained speech and language models, yet most systems remain constrained to monolingual settings and short, isolated utterances. While recent efforts in context-aware ASR…

The ability to follow instructions is crucial for numerous real-world applications of language models. In pursuit of deeper insights and more powerful capabilities, we derive instruction-specific vector representations from language models…

Self-supervised learning (SSL) models have shown exceptional capabilities across various speech-processing tasks. Continuous SSL representations are effective but suffer from high computational and storage demands. On the other hand,…

Large Language Models (LLMs) exhibit remarkable capabilities across various tasks, yet guiding them to follow desired behaviours during inference remains a significant challenge. Activation steering offers a promising method to control the…