6 papers
A Definition of Good Explanations and the Challenges Explaining LLM Outputs
Louis Mahon, Elliot Ford, Callum Hackett
How to define a good explanation is a long-standing philosophical debate which has found recent renewed interest in the context of AI outputs. Explainability is crucial for AI adop…
On the Existence and Behavior of Secondary Attention Sinks
Jeffrey T. H. Wong, Cheng Zhang, Louis Mahon +3
Attention sinks are tokens, often the beginning-of-sequence (BOS) token, that receive disproportionately high attention despite limited semantic relevance. In this work, we identif…
Robust detection of overlapping bioacoustic sound events
Louis Mahon, Benjamin Hoffman, Logan James +7
We propose a method for accurately detecting bioacoustic sound events that is robust to overlapping events, a common issue in domains such as ethology, ecology and conservation. Wh…
Local Compositional Complexity: How to Detect a Human-readable Messsage
Louis Mahon
Data complexity is an important concept in the natural sciences and related areas, but lacks a rigorous and computable definition. In this paper, we focus on a particular sense of…
-TCVAE: On the relationship between Disentanglement and Diversity
Cristian Meo, Louis Mahon, Anirudh Goyal +1
While disentangled representations have shown promise in generative modeling and representation learning, their downstream usefulness remains debated. Recent studies re-defined dis…
Towards a Universal Method for Meaningful Signal Detection
Louis Mahon
It is known that human speech and certain animal vocalizations can convey meaningful content because we can decipher the content that a given utterance does convey. This paper expl…