Introduces Trust-RAG Compass framework and TRC Bench benchmark to assess RAG trustworthiness across factuality, robustness, fairness, transparency, accountability, and privacy, with evaluations showing performance gaps between LLMs.
A Multiscale Visualization of Attention in the Transformer Model
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
abstract
The Transformer is a sequence model that forgoes traditional recurrent architectures in favor of a fully attention-based approach. Besides improving performance, an advantage of using attention is that it can also help to interpret a model by showing how the model assigns weight to different input elements. However, the multi-layer, multi-head attention mechanism in the Transformer model can be difficult to decipher. To make the model more accessible, we introduce an open-source tool that visualizes attention at multiple scales, each of which provides a unique perspective on the attention mechanism. We demonstrate the tool on BERT and OpenAI GPT-2 and present three example use cases: detecting model bias, locating relevant attention heads, and linking neurons to model behavior.
citation-role summary
background 2
citation-polarity summary
verdicts
UNVERDICTED 7roles
background 2polarities
background 2representative citing papers
Induction heads, which implement pattern completion in attention, develop at the same training stage as a sudden rise in in-context learning, providing evidence they are the primary mechanism for in-context learning in transformers.
Performers approximate full-rank softmax attention in Transformers via FAVOR+ random features for linear complexity, with theoretical guarantees of unbiased estimation and competitive results on pixel, text, and protein tasks.
CA-LIG is a unified hierarchical attribution method that computes layer-wise Integrated Gradients fused with class-specific attention gradients to generate signed, context-sensitive explanations for transformer models.
A self-supervised pre-trained neural network with multi-domain channel embedding and self-attention is proposed to create realistic wireless channel models combining deterministic accuracy and stochastic uniformity.
TIDE augments standard transformers with per-layer token embedding injection via an ensemble of memory blocks and a depth-conditioned router to mitigate rare-token undertraining and contextual collapse.
DA-Cramming inserts chunk-level dependency agreement embeddings into a dual-stage pretraining pipeline and reports better downstream performance than prior Cramming baselines.
citing papers explorer
-
Trustworthiness in Retrieval-Augmented Generation Systems: A Survey
Introduces Trust-RAG Compass framework and TRC Bench benchmark to assess RAG trustworthiness across factuality, robustness, fairness, transparency, accountability, and privacy, with evaluations showing performance gaps between LLMs.
-
In-context Learning and Induction Heads
Induction heads, which implement pattern completion in attention, develop at the same training stage as a sudden rise in in-context learning, providing evidence they are the primary mechanism for in-context learning in transformers.
-
Rethinking Attention with Performers
Performers approximate full-rank softmax attention in Transformers via FAVOR+ random features for linear complexity, with theoretical guarantees of unbiased estimation and competitive results on pixel, text, and protein tasks.
-
Explainable AI: Context-Aware Layer-Wise Integrated Gradients for Explaining Transformer Models
CA-LIG is a unified hierarchical attribution method that computes layer-wise Integrated Gradients fused with class-specific attention gradients to generate signed, context-sensitive explanations for transformer models.
-
Realistic Channel Models Pre-training
A self-supervised pre-trained neural network with multi-domain channel embedding and self-attention is proposed to create realistic wireless channel models combining deterministic accuracy and stochastic uniformity.
-
TIDE: Every Layer Knows the Token Beneath the Context
TIDE augments standard transformers with per-layer token embedding injection via an ensemble of memory blocks and a depth-conditioned router to mitigate rare-token undertraining and contextual collapse.
-
DA-Cramming: Enhancing Cost-Effective Language Model Pretraining with Dependency Agreement Integration
DA-Cramming inserts chunk-level dependency agreement embeddings into a dual-stage pretraining pipeline and reports better downstream performance than prior Cramming baselines.