相关论文: A Timeline and Analysis for Representation Plastic…

Mechanistic Interpretability (MI) has emerged as a vital approach to demystify the opaque decision-making of Large Language Models (LLMs). However, existing reviews primarily treat MI as an observational science, summarizing analytical…

Growing concerns over the lack of transparency in AI, particularly in high-stakes fields like healthcare and finance, drive the need for explainable and trustworthy systems. While Large Language Models (LLMs) perform exceptionally well in…

Large language models (LLMs) with Chain-of-Thought (CoT) reasoning have achieved strong performance across diverse tasks, including mathematics, coding, and general reasoning. A distinctive ability of these reasoning models is…

Large Language Models (LLMs) have emerged as a promising cornerstone for the development of natural language processing (NLP) and artificial intelligence (AI). However, ensuring the robustness of LLMs remains a critical challenge. To…

Large Language Models (LLMs) have exhibited remarkable performance across various downstream tasks, but they may generate inaccurate or false information with a confident tone. One of the possible solutions is to empower the LLM confidence…

Recent advances in mechanistic interpretability have revealed that large language models (LLMs) develop internal representations corresponding not only to concrete entities but also distinct, human-understandable abstract concepts and…

Role-playing has emerged as an effective technique for enhancing the reasoning capabilities of large language models (LLMs). However, existing methods primarily rely on prompt engineering, which often lacks stability and interpretability.…

Inference-time intervention (ITI) has emerged as a promising method for steering large language model (LLM) behavior in a particular direction (e.g., improving helpfulness) by intervening on token representations without costly updates to…

Large Language Models (LLMs) often generate inconsistent responses when prompted with semantically equivalent paraphrased inputs. Recently, activation steering, a technique that modulates LLMs' behaviours by adjusting their latent…

Large language models have simplified the production of personalized translations reflecting predefined stylistic constraints. However, these systems still struggle when stylistic requirements are implicitly represented by a set of…

Plasticity loss, a diminishing capacity to adapt as training progresses, is a critical challenge in deep reinforcement learning. We examine this issue in multi-task reinforcement learning (MTRL), where higher representational flexibility is…

Inference-time steering aims to alter a large language model's (LLM's) responses without changing its parameters, but a central challenge is identifying the internal modules that most strongly govern the target behavior. Existing approaches…

Large Language Model (LLM)-based agents are increasingly adopted in high-stakes settings, but current benchmarks evaluate mainly whether a task was completed, not how. We introduce Procedure-Aware Evaluation (PAE), a framework that…

Current safety evaluations of language models rely on benchmark-based assessments that may miss localized vulnerabilities. We present RepIt, a simple and data-efficient framework for isolating concept-specific representations in LM…

Alignment of Large Language Models (LLMs) is the ability to satisfy desired objectives during generation, which is critical for trustworthy deployment. In practice, alignment is often operationalized through multiple objectives such as…

Activation steering is a practical post-training model alignment technique to enhance the utility of Large Language Models (LLMs). Prior to deploying a model as a service, developers can steer a pre-trained model toward specific behavioral…

Recent advancements in large reasoning models (LRMs) have significantly enhanced language models' capabilities in complex problem-solving by emulating human-like deliberative thinking. However, these models often exhibit overthinking (i.e.,…

There is an increasing interest in leveraging Large Language Models (LLMs) for managing structured data and enhancing data science processes. Despite the potential benefits, this integration poses significant questions regarding their…

Steering, or direct manipulation of internal activations to guide LLM responses toward specific semantic concepts, is emerging as a promising avenue for both understanding how semantic concepts are stored within LLMs and advancing LLM…

Aligning large language models (LLMs) with human objectives is crucial for real-world applications. However, fine-tuning LLMs for alignment often suffers from unstable training and requires substantial computing resources. Test-time…