4 citations · 6 across the 10 of their papers we have counts for
7 papers · 1 filter
ShapShift: Explaining Model Prediction Shifts with Subgroup Conditional Shapley Values
Tom Bewley, Salim I. Amoukou, Emanuele Albini +2
Changes in input distribution can induce shifts in the average predictions of machine learning models. Such prediction shifts may impact downstream business outcomes (e.g. a bank's…
Cache What Lasts: Token Retention for Memory-Bounded KV Cache in LLMs
Ngoc Bui, Shubham Sharma, Simran Lamba +2
Memory and computation remain core bottlenecks in long-horizon LLM inference due to the quadratic cost of self-attention and the ever-growing key-value (KV) cache. Existing strateg…
To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models
Anna Hedström, Salim I. Amoukou, Tom Bewley +2
We introduce Mechanistic Error Reduction with Abstention (MERA), a principled framework for steering language models (LMs) to mitigate errors through selective, adaptive interventi…
Interpreting Language Reward Models via Contrastive Explanations
Junqi Jiang, Tom Bewley, Saumitra Mishra +2
Reward models (RMs) are a crucial component in the alignment of large language models' (LLMs) outputs with human values. RMs approximate human preferences over possible LLM respons…
Global Graph Counterfactual Explanation: A Subgraph Mapping Approach
Yinhan He, Wendy Zheng, Yaochen Zhu +5
Graph Neural Networks (GNNs) have been widely deployed in various real-world applications. However, most GNNs are black-box models that lack explanations. One strategy to explain G…
Quantifying Prediction Consistency Under Fine-Tuning Multiplicity in Tabular LLMs
Faisal Hamman, Pasan Dissanayake, Saumitra Mishra +2
Fine-tuning LLMs on tabular classification tasks can lead to the phenomenon of fine-tuning multiplicity where equally well-performing models make conflicting predictions on the sam…
Cross-Domain Graph Data Scaling: A Showcase with Diffusion Models
Wenzhuo Tang, Haitao Mao, Danial Dervovic +4
Models for natural language and images benefit from data scaling behavior: the more data fed into the model, the better they perform. This 'better with more' phenomenon enables the…