相关论文: Towards Reliable Evaluation of Behavior Steering I…

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…

The influence of personas on Large Language Models (LLMs) has been widely studied, yet their direct impact on performance remains uncertain. This work explores a novel approach to guiding LLM behaviour through role vectors, an alternative…

Recent advances in Large Language Models (LLMs) have opened new perspectives for automation in optimization. While several studies have explored how LLMs can generate or solve optimization models, far less is understood about what these…

Multimodal Large Language Models (MLLMs) are evaluated on various benchmarks, such as image captioning, visual question answering, and reasoning. However, many of these benchmarks include overly simple or uninformative samples, complicating…

Despite potential benefits in Software Engineering (SE), adoption of software modelling in industry is low. Technical issues such as tool support have gained significant research before, but individual guidance and training have received…

Understanding what defines a good representation in large language models (LLMs) is fundamental to both theoretical understanding and practical applications. In this paper, we investigate the quality of intermediate representations in…

Large language models (LLMs) exhibit impressive capabilities in generation tasks but are prone to producing harmful, misleading, or biased content, posing significant ethical and safety concerns. To mitigate such risks, representation…

We introduce a machine-learning (ML) framework for high-throughput benchmarking of diverse representations of chemical systems against datasets of materials and molecules. The guiding principle underlying the benchmarking approach is to…

A common problem in LLM evaluation is how to choose a subset of metrics from a full suite of possible metrics. Subset selection is usually done for efficiency or interpretability reasons, and the goal is often to select a ``representative''…

In this paper we discuss recent developments in econometrics that we view as important for empirical researchers working on policy evaluation questions. We focus on three main areas, where in each case we highlight recommendations for…

Sequential multi-agent systems built with large language models (LLMs) can automate complex software tasks, but they are hard to trust because errors quietly pass from one stage to the next. We study a traceable and accountable pipeline,…

Reports of human-like behaviors in foundation models are growing, with psychological theories providing enduring tools to investigate these behaviors. However, current research tends to directly apply these human-oriented tools without…

The predominant method for evaluating the quality of causal models is to measure the graphical accuracy of the learned model structure. We present an alternative method for evaluating causal models that directly measures the accuracy of…

Despite the increasing effectiveness of language models, their reasoning capabilities remain underdeveloped. In particular, causal reasoning through counterfactual question answering is lacking. This work aims to bridge this gap. We first…

Despite advances in large language models (LLMs) on reasoning and instruction-following tasks, it is unclear whether they can reliably produce outputs aligned with a variety of user goals, a concept called steerability. Two gaps in current…

Evaluating the performance of Large Language Models (LLMs) is a critical yet challenging task, particularly when aiming to avoid subjective assessments. This paper proposes a framework for leveraging subjective metrics derived from the…

The growing need for trustworthy machine learning has led to the blossom of interpretability research. Numerous explanation methods have been developed to serve this purpose. However, these methods are deficiently and inappropriately…

The degradation of sewer pipes poses significant economical, environmental and health concerns. The maintenance of such assets requires structured plans to perform inspections, which are more efficient when structural and environmental…

Research and development on conversational recommender systems (CRSs) critically depends on sound and reliable evaluation methodologies. However, the interactive nature of these systems poses significant challenges for automatic evaluation.…

Explanation is a fundamentally human process. Understanding the goal and audience of the explanation is vital, yet existing work on explainable reinforcement learning (XRL) routinely does not consult humans in their evaluations. Even when…